64 results about "Identification scheme" patented technology

The invention relates to the technical field of electroencephalography, in particular to a sleep staging automatic identification method and system based on multi-mode electroencephalography, and the method comprises the following steps: extracting a waveform period to construct dominant distribution, correcting label boundary positioning mutation response, unifying a time axis to form an alignment structure, clustering mutation fragments to establish an alternating relation, and identifying the sleep staging based on multi-mode electroencephalography. And identifying a main channel range to construct an automatic identification scheme. According to the method, a reference channel is positioned through area distribution difference, a boundary change section is judged in combination with a frequency band energy change trend, a channel region with dense and stable mutation points is extracted as an anchor point, time axis unified adjustment is completed according to main response starting and ending time, and an alternating fragment graph is further constructed through a cross-channel time coverage relation of the mutation points. The sequential structure coordination ability and mutation form aggregation expression efficiency among multi-source signals are improved, and accurate recognition of boundary drift and asynchronous response and dynamic extraction of steady-state rhythm in sleep stage division are achieved.

The invention discloses a multiphase flow real-time flow pattern recognition method based on updip pipe pressure and a lightweight deep neural network model, adopts updip pipe pressure data, designs a training strategy, proposes a YI-Net network model, belongs to the field of deep learning, and is used for flow pattern recognition. The method comprises the steps of data preprocessing, model reasoning, loss calculation, back propagation optimization of model parameters and the like. In the data preprocessing process of the training stage, the generalization ability of the model is improved through a data enhancement training strategy, such as Gaussian noise addition, amplitude scaling, overall time translation and the like. The YI-Net model comprises two parts of a backbone network (backbone) and a head (head), and various one-dimensional calculation operations such as one-dimensional convolution and the like are adopted. And the backbone is used for extracting the characteristics of the flow pattern sample. And the head is a Classify layer and is used for determining the category of the backbone output characteristics. The flow pattern recognition scheme provided by the invention has the characteristics of good real-time performance and small parameter quantity, and provides a solution for the field of flow pattern recognition.

The invention relates to the technical field of ore recognition, in particular to a method and a system for quickly and preliminarily recognizing field ore quality of coal-series trilean metal mineral products, which comprises the following steps of: acquiring ore samples and acquiring main peaks, structures and infrared images, extracting image features to generate mutation numbers, analyzing sampling coordinates to reconstruct a connection structure, and calculating the quality of the coal-series trilean metal mineral products. Drawing a closed boundary to judge coordinate attribution, and outputting an ore quality identification scheme. According to the method, main peak reflection, microscopic image layers and infrared spectrum acquisition results are numbered in a unified manner and are subjected to time sequence binding, a synchronous corresponding relation among multi-source images is established, continuous change nodes and sudden change areas are identified in combination with offset tracks and structural features, spatial concentration blocks are generated by using number and coordinate mapping, and a connected structure is reconstructed; and drawing a closed boundary according to the track to form a region classification basis, verifying a mass change direction through image sequence coherence, and outputting a tagging identification result, thereby realizing rapid positioning and accurate classification of the field ore sample.

The invention relates to the technical field of defect identification, in particular to a braided tube appearance defect detection method and system based on machine vision, and the method comprises the following steps: obtaining an image to generate a brightness fracture region, calculating a peak difference to generate a structure abnormal region, extracting a light spot path to identify a burr abnormal region, and analyzing a grid texture to generate a deformation identification region. And a defect identification scheme is obtained by combining gray scale and jump. According to the method, a brightness fracture area is extracted through gray difference symbol change, segmentation of image jump features is realized, a structure dislocation area is positioned through symmetrical wave crest difference, a form mutation zone is revealed through light spot extension and edge consistency comparison, and burr identification stability is improved. Local trend abnormity is judged through texture length and cross frequency quantification, and a multi-dimensional standard is constructed by combining gray span and disturbance frequency screening, so that defect features are clearly presented in a detail level, and the recognition precision and stability in a complex surface environment are ensured.

The invention discloses an open source environment software hidden vulnerability patch identification method, which is based on a multi-stage architecture and collaborative relationship modeling, and provides a new patch group identification scheme: firstly, obtaining candidate code submission from an open source warehouse, extracting correlation characteristics of vulnerabilities and candidate code submission, including rule-based characteristics and semantic characteristics, calculating a correlation score and screening high-correlation submission; pairwise pairing the high-correlation submissions, and fusing multi-dimensional features to predict a cooperative relationship between the submissions; and finally, constructing an undirected graph based on the correlation score, dividing a maximum connected sub-graph, fusing the features in the group through maximum pooling, calculating the correlation with the vulnerability, and outputting an optimal patch group. Meanwhile, an existing patch identification technology based on sorting learning is combined, an enhancement method based on a submission cooperative relation is provided, ranking logic submitted by candidate codes is updated through internal association between code submission and by means of correlation between group vectors and vulnerabilities, and the identification precision in a multi-patch scene is improved.

The invention discloses a five-axis machine tool geometric error identification method based on single-axis and multi-axis measurement, and belongs to the field of numerical control machine tool precision detection. Error kinematics and multi-axis motion models are established based on a multi-body system theory, and error identification models of a linear axis and a rotating axis are respectively constructed. On the aspect of a linear axis, an identification method based on measurement of a positioning error and a straightness error is provided, two measurement / identification schemes of single-axis motion and two-axis linkage are designed, and 18 linear axis geometric error parameters related to the position and 3 linear axis geometric error parameters irrelevant to the position are obtained through calculation. In the aspect of a rotating shaft, 12 rotating shaft geometric error parameters related to the position and 4 rotating shaft geometric error parameters irrelevant to the position are obtained on the basis of an existing linear shaft result based on a measurement and identification scheme of a double-ball-rod instrument in three directions and six positions. According to the method, positioning and straightness error detection are completed in combination with auxiliary equipment such as a laser interferometer and a double-ball-rod instrument, and the method has good universality.

The invention provides a logistics management system and method and a storage medium, and belongs to the technical field of logistics management, and the method specifically comprises the steps: carrying out the determination of an updating recognition scheme of a license plate in the logistics management system according to the license plate data of temporary authority in different dates, on the basis of the updating identification scheme and the similar condition of the license plate and other license plates in the logistics management system, determining the updated required license plate in the license plates, and updating the similar condition between the required license plates and the similar license plate data of the required license plates in the logistics management system; and the permission adjustment strategy for updating the required license plate is determined, so that the security of license plate management is improved.

The embodiment of the invention discloses a target recognition method, device and equipment, a medium and a computer program product. The method comprises the steps of obtaining continuous multi-frame point cloud data corresponding to each target object in a target space; dividing the point cloud data corresponding to the target range according to the center of the point cloud data of each target object in each frame; based on the point cloud features corresponding to the divided point cloud data, generating target features corresponding to each target object in each frame; and performing target identification processing on each target object based on the target features to obtain a target identification result. According to the target identification scheme provided by the invention, the point cloud is associated with the target object, so that each target has exclusive point cloud information, identification is carried out according to the point cloud information of each target, the effect of tracking and identifying each target in the space is realized, multiple targets in the space are accurately identified, and the accuracy of target identification is improved.

The invention discloses a multi-industrial equipment-oriented ELM fault identification method fusing feature attention and manual LU decomposition, and belongs to the technical field of industrial equipment fault diagnosis. The core of the method is that a feature attention mechanism and a manual LU decomposition module are fused: firstly, the feature contribution degree is quantified through a channel attention mechanism, key fault features are automatically focused, weak correlation feature interference is inhibited, and the quality of a hidden layer output matrix is improved; secondly, the manual LU decomposition module realizes autonomous controllability and numerical stability of matrix inverse operation through row principal component selection and regularization optimization, and is separated from the dependence of a third-party math library. According to the method, the accuracy of equipment fault identification is remarkably improved, the omission ratio and the false alarm rate are effectively reduced, memory limitation of edge equipment is overcome, and an efficient and reliable real-time identification scheme is provided for bearing abrasion and gearbox abnormity of a wind power variable pitch system and faults of other industrial equipment such as a machine tool and an industrial motor.

This invention discloses a method, system, and medium for measuring stress and earth pressure behind retaining walls / piles under localized overload. Addressing the shortcomings of existing technologies in simultaneously considering stress or earth pressure measurement under localized overload and rotational displacement of retaining walls / piles, this invention first solves the problem of measuring additional vertical stress behind retaining walls / piles under localized overload. The calculation model is constructed by combining theoretical derivation of the mechanical response of a semi-infinite elastic body surface subjected to a concentrated normal force with experimental test data analysis. Secondly, the invention solves the problem of measuring earth pressure behind retaining walls / piles under localized overload. The optimized scheme provides identification schemes for eight measurement conditions under five types of rotational displacement modes of retaining walls / piles, as well as non-limit active earth pressure measurement schemes under each measurement condition. The measurement method of this invention has a clear calculation model, easily obtainable parameters, and high engineering applicability. This invention also provides a force measurement system based on the measurement method and a medium for carrying the measurement method.

The invention discloses an aquatic organism resource identification and traceability method and system based on image identification and deep learning, and is applied to the technical field of aquatic organism resource traceability. The method comprises the following steps: obtaining an image of an aquatic organism resource, sequentially carrying out ROI region extraction, attitude standardization and key point coordinate extraction and identification, establishing a data file of the whole life cycle of the aquatic organism resource, and storing traceability information to a cloud platform; and assigning a unique identifier to each batch of aquatic organism resources and associating the unique identifier with the traceability information, reading all the traceability information of the corresponding batch of aquatic organism resources through the identification identifier, performing identification through the shot image, and determining whether the identification information is consistent with the traceability information recorded in the traceability information. According to the method, the overall calculation amount is greatly reduced through a staged identification scheme, so that real-time processing in the aquatic organism resource transaction process is easier to realize, and the traceability information is ensured to be real and reliable in combination with a data verification technology.

The invention discloses a large language model intellectual property protection method and device based on domain-specific prejudice feature fingerprints, and belongs to the cross technical field of artificial intelligence safety and intellectual property protection. The method specifically comprises the steps that a three-dimensional orthogonal prejudice data set is constructed, and the data set depicts model prejudice features from the three dimensions of the field, the prejudice type and the response mode; screening the three-dimensional orthogonal prejudice data set based on a two-stage screening mechanism so as to extract core sample data with intra-family high stability and inter-family high discrimination; performing domain-specific prejudice feature fingerprint extraction based on the core sample data to obtain a standardized reference fingerprint; and performing ownership judgment on the to-be-tested model by adopting a confrontation verification mechanism based on the standardized reference fingerprint. According to the method, a quantifiable and high-robustness ownership identification scheme can be provided for the closed-source large language model, and the method has important application value for solving the safety problems of infringement and embezzlement of intellectual property of the large language model and the like.

The utility model provides a light source group, an image receiver, an image acquisition device and electronic equipment, and relates to the technical field of biological recognition, the light source group comprises at least two light sources, each light source is used for emitting polarized light with the polarization degree not equal to 0, and the polarized light emitted by the light source group has at least two polarization directions; the polarized light in at least two polarization directions is reflected by the target object and then carries polarization information corresponding to the contour surface of the target object to sequentially enter the image sensor. The polarized light emitted by each light source in the light source group is optimized, so that the structure of the light source group is simplified, when the light source group is applied to three-dimensional imaging, target object recognition or anti-counterfeiting recognition and other schemes, the schemes can be simplified, and accurate face contour reconstruction and recognition precision improvement are realized with a relatively simple scheme and relatively low cost.

The invention provides a load scheduling processing method and system for a zero-carbon park, and belongs to the technical field of zero-carbon parks, and the method specifically comprises the steps: taking interaction types under different weather types, adjustment matching conditions of a user load in different time periods, and interval time period data of an internet surfing time period as a basis; the method comprises the following steps: determining an update identification scheme of a load scheduling strategy under a weather type when a settled enterprise in a park changes, determining an update processing scheme of the load scheduling strategy under different weather types based on the update identification scheme and a coincidence deviation condition of load data of a newly settled enterprise and an internet surfing time period, and updating the load scheduling strategy under the load scheduling strategy under the different weather types based on the update processing scheme and the coincidence deviation condition of the load data and the internet surfing time period of the newly settled enterprise. And the carbon emission of the park is reduced.

The application belongs to the technical field of image processing, and particularly relates to a crack image segmentation method and system. The robustness of a model to noise is improved by introducing Gaussian noise into image data for enhancement processing. An encoder composed of three layers of progressive stacked residual units is used to extract crack features, and a DropBlock regularization module layer is used to reduce feature redundancy and enhance feature diversity. The image space resolution of the crack features is restored by a decoder symmetrical to the encoder, and the fine structure of the crack boundary is restored. Finally, a dense conditional random field is used to optimize the crack probability distribution map, and the final crack segmentation map is output. The application effectively alleviates the foreground-background class imbalance problem, reduces false positives and false negatives, improves the segmentation performance of the model, and is particularly suitable for deployment on edge devices such as unmanned aerial vehicles and mobile phones, providing a high-precision, low-complexity and strong-robustness crack identification scheme for infrastructure health monitoring.

The invention discloses an aircraft assembly drilling process verification method based on product structural characteristics, and the method comprises the steps: presetting a preassembling scheme based on the product structural characteristics, and adjusting the connection mode and connection spacing of a pre-connection fastener to guarantee the overall stability after preassembling; based on the component assembling scheme, an identification scheme of the pre-assembled fastening nails is reasonably selected and matched; an identification track of the preassembled fastening nail is formulated based on a path optimal principle, and positioning of a product structure form under a global coordinate system is completed; manufacturing indexes of different working condition differences of all hole sites are determined based on aircraft assembly hole manufacturing process correlation parameters; presetting a test working condition based on the manufacturing index of a certain hole site, and enabling the current test working condition to be equivalent to the product working condition; step-by-step optimization is carried out on a hole making process parameter domain based on a test panel test, an optimization result is directionally migrated to product hole making, and process parameter making is completed; and a process strategy is called according to the accumulated information, and hole making of the complete hole making molded surface is completed, so that the correctness of equipment after wire entering is ensured, and the suitability of the process strategy is improved.

This invention defines a symbolic identity and trust protocol for wearable AGI agents, establishing a real-time, consent-anchored passport system that replaces centralized biometric or crypto-based identification schemes. Each passport instance contains a symbolic identity graph linked to cognitive signatures, ethics verifications, and mission context. Consent tokens are issued and revoked dynamically, with real-time swarm synchronization across agents and devices. All interactions—including memory access, network sync, and sensor control—require verification via these passports, enforced through zero-knowledge proofs and treaty-aligned arbitration. The system fully prevents the emergence of exploitative, opaque, or surveillance-based identity regimes like Worldcoin or Meta-linked AGI tokens. Identity becomes sovereign, ethical, and cognitively validated.

This invention discloses a method and system for automated generation and optimization of database table structures. The method specifically includes: receiving the user's backend project source code; extracting a semantic information network graph related to data persistence through static code scanning; inputting the semantic information network graph into an intelligent parsing engine to analyze code logic relationships and obtain index recommendation strategies and potential table structure design flaws; based on the index recommendation strategies and potential table structure design flaws, identifying sensitive fields and recommending encryption or de-identification schemes to obtain a first table structure; performing automated stress testing on the first table structure and adjusting the index strategy and partitioning scheme based on the test results; and optimizing the first table structure according to the adjusted index strategy and partitioning scheme to obtain a second table structure. This invention achieves intelligent generation and optimization of database table structures, improving development efficiency, enhancing maintainability, improving user experience, and ensuring data security.

The application discloses a power distribution network cable joint fault monitoring method and system based on voiceprint analysis, and relates to the technical field of power distribution networks.The method comprises the following steps: performing monitoring urgency evaluation of partial discharge fault monitoring based on predicted material state information and predicted scene state information of a cable joint area in a preset time zone, and setting an adaptive identification time length; taking maximum fault identification accuracy and approximation of the adaptive identification time length as scheme optimization targets, and outputting an optimal fault identification scheme; periodically acquiring original acoustic signals of the cable joint area, and performing signal processing and fault identification on the original acoustic signals according to the optimal fault identification scheme, and outputting a fault monitoring result.The technical problem that the existing cable joint fault identification scheme lacks a dynamic adjustment mechanism, cannot be dynamically adjusted according to real-time equipment states and operating environments, results in low detection efficiency, high missed detection rate and inability to realize real-time monitoring is solved.

The application discloses a geological anomaly body rapid identification method and system based on seismic exploration, and relates to the field of geological anomaly body identification.The method comprises the following steps: fine imaging based on historical seismic data, generation of a high-precision three-dimensional geological model, and construction of a geological regional risk distribution map; emergency evaluation of anomaly body identification, configuration of a plurality of adaptive identification time limits; prediction of the predicted seismic source signal sequence and the predicted noise signal sequence of the second geological region based on the historical seismic source signal sequence and the historical noise signal sequence, analysis of the anomaly body identification complexity, and output of the second optimal identification time limit; optimization of the identification strategy of the geological anomaly body, execution of the geological anomaly body identification of the second geological region, iterative optimization of the identification scheme, and completion of the coal mining task.The geological anomaly body rapid identification method and system solve the problem that the geological anomaly body identification strategy cannot be dynamically adjusted, resulting in low safety warning responsiveness and insufficient accuracy.

The invention provides a partial discharge signal automatic positioning method and system and a storage medium, and belongs to the technical field of signal processing, and the method specifically comprises the steps: determining the partial discharge abnormal data of different switch cabinets according to the detection data of the historical partial discharge data of the switch cabinets, determining an identification scheme of the setting target of the partial discharge online monitoring device based on the partial discharge abnormal data and the potential setting target, and according to the change condition of the setting target and the identification results of the setting target under different partial discharge abnormal types, determining that the reliability degree of monitoring analysis processing does not meet the requirement. According to the method, the monitoring result of the partial discharge exception type in the set target is determined, and the automatic positioning processing method of the exception type is determined in combination with the positioning deviation conditions under different partial discharge exception types, so that the accuracy of positioning processing of the partial discharge source is improved.

This application provides a method, apparatus, device, and medium for identifying and processing recyclable materials. In this method, image data and weight data of the item to be recycled are acquired; the image data are input into a first recognition model and a second recognition model respectively, and inference is performed in parallel to output corresponding material identification results and product identification results; based on the material identification results, product identification results, and weight data, it is determined whether the recyclable item is a qualified recyclable; if it is determined to be a qualified recyclable, a first control command is output, or if it is determined to be an unqualified recyclable, a second control command is output. This solution addresses the problem in related technologies where recyclable material identification schemes cannot meet the requirement of acquiring directly recyclable aluminum cans and determining their product attributes, thus affecting the recycling effect.

The application provides a long-tail distribution object identification scheme fusing an unsupervised domain adaptation technology, comprising an object identification main system, wherein the object identification main system comprises a domain adaptation system, an object library and a model center; the domain adaptation system comprises a prototype level contrast learning and a consistency loss based on a smoothing label; the object library comprises an object image acquisition device and a database; and the model center comprises an object pre-training model and a long-tail identification model. The long-tail distribution object identification scheme can make the photographing object identification module in a shopping / science popularization application program more accurately identify objects in a natural life scene.

The present disclosure provides a method and a system of automatically identifying goods storage and retrieval. A method of dividing key monitoring areas corresponding to entrances and exits and intelligent shelves, activating user positioning and automatically weighing goods can realize intelligent and simple goods storage and retrieval for users without binding Radio Frequency Identification (RFID) tags for each piece of goods, thus greatly reducing consumables and labor cost. At the same time, the method can also avoid the problems of high computing cost and easy misidentification of a machine vision identification scheme. In addition, the storage and retrieval prediction algorithm can quickly determine and broadcast information of the retrieved / stored goods, thus reducing the waiting time of users and improving the user experience.

The application discloses a flow scheduling method and device and a medium, and relates to the technical field of communication. The method applied to a user-side device comprises the following steps: in the case that a first data packet of a data flow is listened to, source identification information of the first data packet is identified, the source identification information comprises user identification information and application identification information; flow identification information is generated, the flow identification information comprises the source identification information; the first data packet and the flow identification information are sent to a network-side device, so as to instruct the network-side device to, in the case that a second data packet of the data flow is received and five-tuple information of the second data packet is the same as five-tuple information of the first data packet, perform flow scheduling on the second data packet according to the source identification information. The application can improve the protocol applicability and compatibility of the flow identification scheme.

The invention relates to blind identification of Turbo codes in a non-cooperative communication environment, and provides a random interleaving mode identification method based on interleaving Turbo codes. The method comprises the following steps: firstly, realizing interleaving mode identification of an outer interleaver and an inner interleaver on the basis of a Turbo code coding system added with the outer interleaver; secondly, an LD-based identification scheme is proposed for the first time in the scene, and a KL divergence-based scheme is further proposed to identify the interleaver in the system, so that compared with the LD-based scheme, the influence of variance is considered, and the performance is further improved; moreover, aiming at the condition that only one RSC code is successfully recovered, a scheme of secondary identification after forward updating is adopted, so that the identification accuracy is further improved. The random interleaving mode identification of the interleaved Turbo code under the low signal-to-noise ratio is realized.

The invention provides a premixed concrete quality detection method and device, a medium and a product, and relates to the technical field of image processing, and the method enables a target image detection model to capture and enhance time sequence change information related to concrete internal flow from an input image sequence through a space-time dynamic sensing module and a space-time attention module. Characteristics containing dynamic information are purified on different scales through a channel-space double attention module, and key apparent characteristics under each scale are focused. And finally, the adaptive feature pyramid network dynamically adjusts the fusion strategy according to the image content, and performs intelligent weighted fusion on the multi-scale dynamic features. The mode of combining time sequence dynamic analysis and self-adaptive multi-scale space analysis enables the target image detection model to form more comprehensive identification on the state of the concrete, and the technical problem that a concrete state identification scheme is poor in identification effect under complex and dynamic actual working conditions in a transportation tank car is solved.

The invention discloses a fire source recognition method, system and device based on multi-feature fusion and a storage medium, and relates to the technical field of fire fighting, the fire source recognition method based on multi-feature fusion comprises the steps that a visible light image and thermal radiation data of a target area are collected, and the visible light image is collected by a binocular camera; determining that a fire source exists in the target area under the condition that the flame area identified in the visible light image is matched with the temperature gradient map corresponding to the thermal radiation data and the preset ultraviolet signal is detected in the target area; and performing binocular vision positioning on the fire source according to the visible light image to obtain a three-dimensional space coordinate of the fire source. The technical problem that a traditional fire source recognition scheme is prone to being interfered by the environment, and false alarm is caused is solved.

The invention provides a test board card management method, device, system and equipment, and the method comprises the steps: determining a corresponding relation between a server and a test board card through a system bus scanning operation, and carrying out the scanning operation based on the corresponding relation between a slot position and a test channel; the method comprises the following steps: controlling a test channel corresponding to each slot to perform level switching and detecting an output level change condition of each test board card to determine a corresponding relationship between the slot and the test board card, thereby generating a target corresponding relationship table capable of representing the corresponding relationship among a server, the slot and the test board card; compared with a mode of manually checking and recording the mapping relation in the prior art, an automatic identification scheme for the mapping relation in combination with software operation and hardware electric signals is provided, automation of mapping relation confirmation can be achieved, the labor cost is reduced, the accuracy of configuration records is improved, an unsupervised batch configuration mode can be supported, and the user experience is improved. The method is suitable for a large-scale production environment.

This application relates to the field of terminal technology, providing a terminal mode switching method, a mobile terminal, and a computer-readable storage medium. The method includes: the mobile terminal determining that it is in a first terminal mode based on motion data collected by a motion sensor; and the mobile terminal determining that it is in a second terminal mode based on changes in the duration of its switching access to various cellular cells; if the first terminal mode is a non-high-speed rail mode and the second terminal mode is a high-speed rail mode, the mobile terminal enters the high-speed rail mode. Therefore, this method determines whether the mobile terminal has entered the high-speed rail mode based on two sets of high-speed rail identification schemes. By considering multiple perspectives, it can reduce reliance on sensors to a certain extent, thereby improving the accuracy of high-speed rail mode identification and ensuring the smooth implementation of communication schemes on the modem that rely on the high-speed rail mode.