35 results about "Label free" patented technology

The invention discloses a semi-supervised image annotation method based on a limited label data set, and the method comprises the steps: taking a FixMatch frame as a basis, and integrating a learnable batch normalization module, a dual-scale parallel convolution module, a content and style separation dual-branch module and a dynamic residual gating module in a ResNet backbone network, the stability of feature extraction and the adaptive capacity to enhanced disturbance are improved. For a label-free sample, a multi-level pseudo-label fusion mechanism is provided, prediction distribution of weak, medium and strong enhanced views is synthesized, and high-confidence pseudo-labels are generated through confidence weighted fusion of multi-level enhanced views and comparison and screening with a category threshold. On the basis, a joint loss function composed of label supervision loss and pseudo label consistency loss is constructed, and a plurality of key control parameters in FixMatch + + are adjusted and optimized in a pre-experiment and grid search combined mode to obtain a group of optimal parameters of the model. Finally, a user inputs a label-free image into the trained FixMatch + + model, and the model can automatically generate a high-confidence pseudo label, so that the number of labeled samples in a limited labeled image set is increased, and the classification precision is improved. By implementing the method, the manual annotation cost can be reduced, and efficient and reliable support is provided for image analysis and recognition tasks.

The invention relates to the field of three-dimensional vision, in particular to a label-free three-dimensional point cloud segmentation method based on a visual large model, which comprises the following steps: acquiring three-dimensional point cloud data, and extracting a boundary point set according to the three-dimensional point cloud data in combination with curvature and normal vector to construct a target function; solving the objective function by using an optimization algorithm, selecting an optimal projection visual angle combination, and extracting two-dimensional features of all two-dimensional projection images and three-dimensional features of three-dimensional point cloud data by using a visual large model; performing consistency matching on the two-dimensional features and the three-dimensional features through a matching algorithm to obtain a plurality of matching results; acquiring a definition weight factor and a coverage rate weight factor to calculate an importance weight; fusing the two-dimensional features according to the weight factors to obtain overall fused two-dimensional features; and inversely mapping the integrally fused two-dimensional features to a three-dimensional space to realize three-dimensional point cloud unmarked segmentation. The three-dimensional point cloud segmentation method has the effect of realizing three-dimensional point cloud automatic segmentation without labels.

A particulate matter detection device takes holographic images of flowing particulate matter concentrated by a virtual impactor, which selectively slows down and guides larger particles to fly through an imaging window. The flowing particles are illuminated by a pulsed laser diode, casting their inline holograms on a CMOS image sensor in a lens-free mobile imaging device. The illumination contains three short pulses with a negligible shift of the flowing particle within one pulse and triplicate holograms of the same particle are recorded at a single frame revealing different perspectives of each particle. A deep neural network classifies the particles based on the acquired holographic images. The device was tested using different types of pollen and achieved a blind classification accuracy of 92.91%. This mobile and cost-effective device weighs ˜700 g and can be used for label-free sensing and quantification of various bio-aerosols over extended periods.

The application belongs to the field of artificial intelligence model compression and edge deployment in intelligent operation and maintenance and fault diagnosis of power equipment, and discloses a transformer cooperative distillation incremental learning method and system based on time sequence consistency, which comprises the following steps: a teacher model is used to screen a no-label sample set to obtain a pseudo-label sample set, the pseudo-label sample set is combined with an original label sample set to obtain a distillation training sample set; a multi-mechanism cooperative distillation training scheme containing soft and hard label joint distillation, time sequence consistency distillation and multi-task distillation is constructed, and a student model is subjected to distillation training through the multi-mechanism cooperative distillation training scheme; and a cooperative distillation incremental learning scheme in which a cloud end continuously updates a teacher model to adapt to new multi-source monitoring data and a student model after edge distillation generates a pseudo-label to expand the distillation training sample set is constructed. The application can be used for realizing high-precision and low-cost training and continuous updating of a student model in a resource-limited terminal or an online scene.

The invention provides a few-label industrial process quality prediction method based on a DCMSViT model, and relates to the technical field of industrial process monitoring, and the method comprises the steps: carrying out the preprocessing of historical data, and dividing the historical data into a labeled sample set and an unlabeled sample set; then, a dynamic collaborative regression module is utilized to generate pseudo labels for unlabeled samples based on labeled samples, and high-confidence pseudo label samples are screened out through a dynamically calculated confidence threshold, so that a labeled training set is expanded; then, the expanded labeled sample set and the remaining unlabeled sample set are input into a multi-scale visual Transform module together for training, and the multi-scale visual Transform module is used for extracting global feature vectors of the data so as to carry out quality prediction; and finally, the trained DCMSViT model is utilized to realize real-time and online prediction of the key quality variables of the industrial process. According to the method, a small amount of label data and a large amount of label-free data are effectively utilized, and the accuracy of industrial process quality prediction and the model generalization ability in a small-label scene are remarkably improved.

The present disclosure relates to a model training method, an instant pushing method, a device, a medium and an electronic device, and belongs to the technical field of computers, and can realize model training in an end intelligent scene. A model training method based on federated learning comprises: multiple first participants with labels, the same features and different samples, encrypting the label values possessed by the multiple first participants and transmitting the encrypted label values to a second participant without labels, the second participant possessing all the samples of the first participants but having different features from the first participants; the second participant determining a second gradient based on the encrypted label values and feature values of the features possessed by the second participant and transmitting the second gradient to the first participants; the first participants determining a first gradient and a first histogram based on the feature values of the features possessed by the first participants and the label values possessed by the first participants and transmitting the first gradient and the first histogram to the second participant; the first participants determining a first optimal split point based on the second gradient and transmitting the first optimal split point to the second participant; and the second participant determining a global optimal split point based on the first gradient, the first histogram and the first optimal split point and transmitting the global optimal split point to the first participants.

The invention relates to a vibration meter error monitoring and calibrating method and system based on data analysis, and the method comprises the steps: obtaining a plurality of paired original vibration signals, and screening and marking credible samples and suspicious samples; calculating a first error value of the credible sample, generating a first training set with a credible label according to the first error value, and training a first error model; predicting the suspicious sample according to the first error model, and generating a pseudo label; calculating the prediction uncertainty of a pseudo label, screening a pseudo label sample, combining the pseudo label sample with the first training set, generating a second training set, and training a second error model; inputting the vibration signal to be calibrated into the second error model to obtain a real-time error; and the real-time error is calibrated. According to the technical scheme, the true label is generated by using the high-credibility sample, a large amount of label-free field data is converted into available training data by means of pseudo label self-training, the model accuracy is improved, and therefore the error monitoring and calibration precision of the vibration meter is improved.

The application discloses a pseudo-label model training method and device, a storage medium and electronic equipment. The method comprises the following steps: obtaining a plurality of sample communication relationship data with added sample labels; using a first classification model to analyze the sample communication relationship data, and generating a first classification label, wherein the first classification model is trained using clean label data; using a second classification model to analyze the sample communication relationship data, and generating a second classification label, wherein the second classification model is trained using mixed data, and the mixed data comprises clean label data, noise label data and unlabeled data; determining that the difference between the sample label and the second classification label is a first difference, and the difference between the first classification label and the second classification label is a second difference; and using a preset loss function model to analyze the first difference and the second difference, and determining a pseudo-label model. The application solves the technical problem that unlabeled data cannot be used for model training.

The embodiment of the invention provides a robot control method, electronic equipment, a storage medium and a program product, a UWB sensing module is deployed on a robot, the robot control method comprises the steps that the current mode of the UWB sensing module is determined, the current mode is a label mode or a label-free mode, the UWB sensing module senses a UWB label in the label mode, and the UWB label is sent to the electronic equipment; in the no-label mode, the UWB sensing module does not sense the UWB label; determining a first interaction behavior according to the current mode and the collected data of the UWB sensing module; and executing the first interaction behavior. Therefore, the accuracy of the interactive behavior decision of the robot is improved.

The application discloses a kind of label-free inference enhancement methods based on generator-verification collaborative evolution, which utilizes large language model as generator and verifier simultaneously: first, generator generates result set multiple times on inference query inference, and drafts pseudo-label to divide positive and negative sample groups through majority vote;Subsequently, the verifier verifies the positive sample to form a positive verification set, and generates a verification result for the negative sample to form a negative verification set;Based on the positive and negative sets, contrast training samples are constructed to correct errors and optimize the inference and verification capabilities of the model. The optimized model can be directly used for inference and verification of target queries. Experiments show that on mathematical reasoning and cross-domain question answering datasets, this method significantly improves inference and verification accuracy, while training stability is superior and has outstanding cross-distribution generalization capability.

The application relates to a remote sensing image cross-domain migration classification method for a label-free target area, comprising the following steps: acquiring multi-source remote sensing data covering a region; constructing an adversarial domain self-adaptive network; calculating a classification loss and a domain discrimination loss; constructing an adversarial training total loss function, and reversely propagating network parameters after dynamically adjusting a weight coefficient to update the network parameters; inputting target domain images into a trained feature extractor and a classifier, and finally outputting a refined classification result map. The application has the beneficial effects that the application enhances the multi-dimensional representation capability of a model for a complex geographical environment, overcomes the limitation of a single data source, and provides a more robust feature basis for cross-domain migration classification.

The application belongs to the field of electrochemistry, and relates to tumor protein P53 gene detection, in particular to a label-free double-color test strip, a kit and a preparation method thereof for tumor protein P53 gene detection. The application belongs to a novel LFA method, which combines DNA hybridization chain reaction with the electrostatic adsorption principle of test line / control line, selects common and cost-effective bovine serum albumin as the test line, utilizes the difference in the adsorption capacity of BSA to double-stranded DNA and nano-gold, and the different flow rates of different molecules on the NC membrane to complete the detection of p53 at different concentrations. The test line is constructed by positively charged PDDA to ensure the effectiveness of the LFA. Meanwhile, the detection principle of a liquid colorimetric sensor is combined, and the double-color development greatly reduces the detection limit of the test strip. The sensor is label-free throughout, and has high analytical performance in terms of sensitivity, selectivity and practicability. The p53 gene is detected on the lateral flow test strip for the first time.

The invention discloses an underwater sound target identification method and system based on soft label guidance and domain adversarial fine tuning, and the method comprises the steps: employing a labeled underwater sound data set, an unlabeled underwater sound data set, and a source domain general audio data set; a joint loss function including classification loss, soft label distillation loss and domain adversarial loss is constructed to assist the knowledge distillation process of the teacher model to the student model; meanwhile, dynamic convolution and dynamic activation functions are introduced into a dynamic convolution feature extraction module in the student model, so that the adaptive modeling capability of the model to input signal characteristics is enhanced; besides, a multi-layer enhanced long short-term memory network module and a residual connection mechanism are arranged in a feature enhancement module, and an index gating mechanism and a new memory mixing mode are introduced into the enhanced long short-term memory network module, so that the problem of gradient disappearance or explosion in a deep network is solved; and the modeling capability of the long-time sequence dependency relationship is effectively enhanced.

This application discloses a method, apparatus, electronic device, medium, and program product for pseudo-labeling. The pseudo-labeling method includes: obtaining a real label sample set; labeling an unlabeled sample set as an initial pseudo-label sample set based on a first labeling model constructed from the real label sample set; dividing the initial pseudo-label sample set into a first pseudo-label sample set and a second pseudo-label sample set; re-labeling the second pseudo-label sample set using a second labeling model jointly constructed from the real label sample set and the first pseudo-label sample set to obtain a third pseudo-label sample set; calculating the label difference value between the pseudo-labels labeled in the second and third pseudo-label sample sets; and filtering the target pseudo-label sample set by removing samples from the second and third pseudo-label sample sets whose label difference value is greater than a preset label difference threshold. This application solves the technical problem of low accuracy in pseudo-labeling in the prior art.

A method of performing cell assays, comprising: inputting in a computing system a digital refractive index (RI) image of a sample containing a plurality of cells, applying a segmentation algorithm in said computing system to process said digital RI image configured to locate and define an outer boundary of each cell of said plurality of cells, applying in said computing system RI values from the RI image to the segmented cells, calculating in said computing system from the RI values, metrics including any of composition, structure and shape of each segmented cell, evaluating the metrics with a machine learning module in said computing system to classify a physiological state of each of the cells.

The present application relates to the technical field of computer vision and image processing, and discloses a remote sensing weakly supervised fine-grained target detection and recognition method and device, based on a general text prompt corresponding to a small amount of labeled remote sensing data samples and a large amount of unlabeled remote sensing data samples, by extracting the geometric features of the samples and the cross-modal features representing the visual text differences, combining the fine-grained class labels of the labeled remote sensing data samples, prior knowledge prototypes of various fine-grained classes are constructed; then, the prior knowledge prototypes of various fine-grained classes are used to construct fine-grained soft labels of the unlabeled remote sensing data samples, which are used as supervision signals to realize the training of the model, solve the cost consumption problem of fine-grained labeling, improve the label labeling efficiency of the unlabeled remote sensing data samples, and further improve the model training efficiency and target recognition efficiency.

The invention provides a label replacement method and device, electronic equipment and a storage medium, relates to the technical field of big data processing, and can accurately and efficiently replace labels in batches. A first label data set of a first version and a second label data set of a second version are obtained. Each label data set comprises a plurality of pieces of corresponding label data and label information corresponding to each piece of label data, each piece of label data comprises a label value of a fixed label, the label information indicates label value conditions of a plurality of fixed labels in the label data, and the label value conditions comprise label values and label-free values. Deleting marks are added to all label values of the first part data and the third part data in the multiple pieces of first label data, classification marks are added to all label values of the second part data in the multiple pieces of first label data based on first mark information and second mark information, and multiple pieces of reconstructed first label data are obtained; and updating the tags in the database based on the reconstructed first tag data.

The application provides a kind of telecommunication anomaly detection method and detection device.The method comprises: obtaining the social relationship information and portrait feature information of each user in personnel relationship heterogeneous graph, obtaining the fusion feature information and relationship feature information of each user according to label information, portrait feature information and social relationship information;According to fusion feature information and relationship feature information, adopt neighborhood aggregation algorithm, obtain the node feature information of label user;Based on semi-supervised learning mechanism, according to social relationship information and node feature information, through improved attention mechanism model, obtain the anomaly probability of each user, to realize through label propagation algorithm, the label information of label user is propagated to unlabelled user, and anomaly detection is completed.The method of the application utilizes the context information of multiple relationships in heterogeneous graph and the characteristics of semi-supervised learning, effectively improves the detection accuracy and adaptability of the model in the label scarce scene.

The application discloses a semi-supervised image labeling method FixMatch++ based on a limited label data set, which is based on the FixMatch framework, and four modules, including a learnable batch normalization, a double-scale parallel convolution, a content and style separation double branch, and a dynamic residual gate, are integrated in a ResNet backbone network to improve the stability of feature extraction and the adaptability to enhanced perturbations. For unlabeled samples, a multi-level pseudo label fusion mechanism is proposed, the prediction distribution of three types of enhanced views, including weak, medium and strong, is integrated, high-confidence pseudo labels are generated by multi-level enhanced view confidence weighted fusion and comparison with a class threshold, and a joint loss function composed of a labeled supervised loss and a pseudo label consistency loss is constructed. Through pre-experiments combined with grid search, a group of optimal parameters of the model are obtained by adjusting and optimizing a plurality of key control parameters in FixMatch++. Finally, a user inputs an unlabeled image into the trained FixMatch++ model, and the model can automatically generate high-confidence pseudo labels, thereby expanding the number of labeled samples in the limited labeled image set and improving the classification accuracy. The implementation of the method can reduce the cost of manual labeling and provide efficient and reliable support for image analysis and recognition tasks.

Provided are methods and devices for the label free detection of analytes in solution, including analytes suspended in a biological fluid. A field effect transistor (FET) is positioned in close proximity to a paired set of reference electrodes and the reference electrodes electrically biased to provide desalting and a stable gate voltage to the FET. In this manner, charged ions are depleted in the sensing region of the sensor and device sensitivity to analyte detection improved by the removal of charge that otherwise interferes with measurement. Also provided are methods and systems providing increased in reference electrode surface area and / or decrease in droplet volume to further improve label-free detection of analytes.

The invention discloses an unmanned aerial vehicle rotor bearing fault diagnosis method and system based on multi-source data, and relates to the technical field of bearing fault diagnosis, and the method comprises the steps: obtaining a to-be-diagnosed sample during the operation of an unmanned aerial vehicle rotor bearing; extracting a multi-source fusion implicit code feature vector, inputting the multi-source fusion implicit code feature vector into a fault diagnosis learning device, and outputting a fault diagnosis result; the training method of the feature extractor and the fault diagnosis learner comprises the following steps: collecting labeled samples and unlabeled samples; constructing an integrated model; performing expert marking on K unlabeled samples with the highest comprehensive score, and supplementing the K unlabeled samples to a set of labeled samples; predicting a false label for the unmarked label-free sample and obtaining a false label sample; the pseudo label samples are incorporated into a training set, wherein the training set further comprises label samples and marked non-label samples; and training the integrated model. According to the method, the problems of insufficient utilization of multi-source features, serious waste of unlabeled samples and insufficient representativeness of the unlabeled samples are solved.

The invention discloses a label-free visual dynamic target tracking and positioning method for an orchard transportation robot, and relates to the technical field of robot visual perception and intelligent control. According to the method, a depth camera is utilized, a hand raising action is used as a natural interaction mode to trigger a transportation robot to follow, a YOLO model is adopted to detect a first hand raising person, and a KCF and CSRT double-tracker algorithm is adopted; a body direction vector is constructed in combination with the human body key points, and 3D center point coordinates and quaternion postures of the tracked target person are solved; kalman filtering prediction is introduced to assist in re-detection, and follow-up start-stop and target switching are achieved through gesture semantic judgment. According to the method, a target person does not need to wear a label or special equipment, light-weight target detection, multi-modal visual tracking and a dynamic attitude estimation strategy are fused based on a depth camera video stream, natural interactive autonomous following and spatial positioning of orchard operating personnel are realized, and the man-machine collaborative transportation efficiency and the intelligent level under complex terrains are improved.

The present application relates to the technical field of gearbox intelligent operation and maintenance, and particularly relates to a wind turbine gearbox intelligent diagnosis method with few labels, comprising: obtaining source domain and label-free target domain data of wind turbine gearbox with fault labels; obtaining sample pseudo-labels by using a fault diagnosis network; training the fault diagnosis network with a minimized cross-entropy loss function by using labeled source domain samples; training the fault diagnosis network with a minimized unsupervised entropy loss function by using unlabeled source domain samples; training the fault diagnosis network with an enhanced cross-entropy loss function by using labeled and unlabeled source domain samples; and constructing a domain perception prototype generator and training the domain perception prototype generator by using an optimization algorithm and a semantic alignment constraint function. The present application solves the problems of poor model generalization and difficulty in identifying target domain fault categories when the target domain cannot participate in training in the prior art.

The invention provides a cross-central-domain adaptive reasoning method and system based on a condition adapter, a storage medium and electronic equipment, and the method comprises the steps: constructing a lightweight condition adapter module for each medical center on the basis of a shared trunk model of pre-training and freezing parameters; training adapter parameters on the premise that a backbone network is not updated by using non-label data or a small amount of label data of a target domain; in the reasoning stage, a matched adapter is selected from an adapter library according to a source identifier or a feature fingerprint of input data and is embedded into a trunk model, so that dynamic calibration of intermediate features is realized. The method supports that the trunk model is unchanged, only the deployment mode of the adapter is updated, parameter increment is small, reasoning is flexible, change tracing and rollback can be achieved through a version control mechanism, and the method is suitable for safe and controllable deployment of multi-center AI systems such as medical image analysis and pathological recognition.

The invention relates to a plastic container (1) for receiving liquids, in particular beverages, having a base portion (2), a main body (4) which adjoins the base portion (2) and which forms a receiving volume for receiving the liquid and which extends in a longitudinal direction (L) of the plastic container, a shoulder region (8) which adjoins the main body in the longitudinal direction (L) and in which a cross section in the longitudinal direction (L) is reduced, and having a mouth region (10) which adjoins the shoulder region (8) in the longitudinal direction (L) and in which a mouth is formed via which the liquid can be supplied to the container, wherein the shoulder region and the mouth region are arranged in an upper section of the plastic container relative to the longitudinal direction (L), wherein the plastic container has a deformation portion (6) below the mouth portion (10) in the longitudinal direction (L), in which deformation portion the plastic container (1) is elastically deformable with respect to its longitudinal direction and wherein the container is label-free and preferably also printing-free.

A method for an adaptive compression scheme that dynamically adjusts to data characteristics, maintaining model classification accuracy while optimizing compression efficiency, the method including receiving, from an edge device, a sample of compressed data and a sample of raw data that has not been compressed, and the sample of compressed data and the sample of raw data are unlabeled, decompressing the compressed data to obtain decompressed data, and classifying, with an ML (machine learning) model, the decompressed data, using the ML model and the raw data to update a compression quality parameter, and transmitting the compression quality parameter to the edge device, and the compression quality parameter is usable by the edge device to control compression of a subsequent sample of compressed data.

The application discloses a polyp detection model training method based on deep learning, and is characterized in that the method comprises the following steps: training a basic polyp detection model by using source domain report images; extracting pseudo-label samples from a target domain video without labels by using the trained basic polyp detection model, wherein the pseudo-label samples comprise positive pseudo-label samples and hard negative samples; retraining the basic polyp detection model by using source domain report images mixed with the hard negative samples, and fine-tuning the basic polyp detection model by using the positive pseudo-label samples. The technical scheme disclosed by the application improves the accuracy and reliability of polyp detection and the establishment of a polyp data set, and reduces the labeling cost.

Embodiments of the present application provide a label generation method and device, electronic equipment and storage medium, relating to the technical field of machine learning, the method comprising: inputting each unlabeled data into a model trained based on labeled data to obtain a first predicted label representing the probability of the unlabeled data belonging to each preset category; clustering the unlabeled data based on the data features of the unlabeled data and the first predicted label to obtain a first clustering cluster, and clustering the labeled data based on the data features of the labeled data and the true value label to obtain a second clustering cluster; for each first clustering cluster, obtaining a second predicted label representing the probability of the first clustering cluster belonging to each preset category based on the feature distance between the first clustering cluster and each clustering cluster set; and fusing the first predicted label and the second predicted label of each unlabeled data to obtain a fused label and a final label. In this way, effective labels of unlabeled data can be generated.

The utility model relates to label detection technical field discloses a kind of plastic bottle no-label detection mechanism, it includes belt conveyor, the rack of belt conveyor is fixed with workstation, the tabletop of workstation straddles the belt of belt conveyor, and disc is installed on the workstation, disc is driven by reduction motor fixed on the workstation, baffle and support are also fixed on the workstation, baffle is located between disc and support, and scanning code slit is opened in baffle, and code reader is installed on support, and code reader scans the label of plastic bottle by scanning code slit.The utility model has the beneficial effects that: plastic bottle is placed on disc to rotate, label is read by code reader, if code reader is not read-write to label, it indicates that the bottle is missed, it can be rejected, prevent the occurrence of misdiagnosis while reducing worker labor intensity, avoid visual fatigue.

The present application relates to an automatic production equipment for label-free packaging bottles, which belongs to the technical field of packaging bottle production. The equipment comprises a machine frame, wherein a bottle body conveying device and a bottle body labeling device are arranged on the machine frame, wherein the bottle body labeling device comprises a rotating shaft arranged on the machine frame and an driving annular plate coaxially fixed on the rotating shaft, a driving mechanism for driving the rotating shaft to rotate is arranged on the machine frame, a plurality of storage plates are rotatably arranged on a top portion of the driving annular plate, the plurality of storage plates are distributed in a circular array along a circumferential direction of the driving annular plate, at least one self-rotating motor for driving each of the plurality of storage plates to rotate is arranged at a bottom portion of the driving annular plate, a servo annular plate is coaxially fixed on the rotating shaft, pressing assemblies corresponding to the storage plates are arranged on the servo annular plate, a plurality of laser heads for bottle labeling are arranged on the machine frame. The present invention is conducive to meeting the labeling requirements of large breadth and high production capacity for packaging bottles, is conducive to the mass production of the label-free packaging bottles, and is conducive to environmental recycling, energy consumption saving, and production cost reduction.