101 results about "Response area" patented technology

The invention relates to the technical field of data analysis, in particular to a road traffic intelligent operation and maintenance method and system based on multi-source data fusion, and the method comprises the steps: constructing a three-dimensional model through a credibility calculation module by integrating three types of factors, namely a road environment, an equipment state and historical verification, generating a dynamically updated credibility index, and correcting a weight in real time based on feedback; the problem that the reliability of the multi-source data fluctuates dynamically due to the influence of environmental interference, equipment aging and historical deviation is solved; the road feature conflict resolution module maps to feature dimensions corresponding to road topology conflict features through dynamic credibility indexes, constructs a road conflict decision matrix, and fuses and generates a road anomaly decision set with credible labels according to the feature dimensions; the resource scheduling module is used for dividing three-level response areas based on credible indexes, executing accurate scheduling of triple constraints in combination with device priorities and credible labels and generating a resource-decision association graph, and self-adaptive optimal configuration of limited maintenance resources according to data reliability is achieved.

The invention belongs to the technical field of digital twinning, and discloses a hydraulic engineering-oriented digital twinning model dynamic construction and risk assessment method, which comprises the steps of dividing a hydraulic engineering into a static basic component area and a dynamic response area, accessing historical hydraulic engineering data, combining BIM and a finite element analysis technology, and carrying out dynamic construction and risk assessment on a digital twinning model. Constructing a structure model of the static basic component area and an agent model of the dynamic response area; combining the structured model with the proxy model to obtain the hydraulic digital twin; state data in the water conservancy digital twinborn body operation process are collected, and a twinborn state sequence is generated; constructing a water conservancy semantic map, and carrying out semantic binding on the water conservancy digital twin, the real-time water conservancy data and the control logic; dynamically updating a twinning state sequence when the node state of the water conservancy semantic map changes by adopting an event-driven strategy; and the response time efficiency and the emergency regulation and control capability of the water conservancy project are further improved.

The invention discloses an SPR response region identification method based on image semantic segmentation and time sequence alignment, and the method comprises the following steps: collecting SPR image frame sequence data, and constructing an original image sequence; performing image preprocessing operation on the original image sequence, and outputting a standardized image sequence; constructing a time sequence window image set composed of multiple continuous frames; inputting the time sequence window image set into an improved SegFormer model, and generating a response region segmentation mask image corresponding to each frame; executing cross-frame time sequence alignment operation of the response area, and outputting time sequence consistency identification mapping of the response area; performing area statistics, intensity analysis and time positioning operation; and generating a structured response region recognition result. The spatial-temporal evolution process of the response area in the SPR image sequence can be effectively recognized, the accuracy and stability of response area recognition are improved, and the method is suitable for high-precision biological detection and real-time molecular analysis scenes.

The invention relates to a method for setting power loss of a mountain photovoltaic array. The method comprises the following steps of: establishing a coupling causal spectrum of multi-source factors including terrain, angle, sunlight, temperature, dust and wiring mode to component output; identifying a dominant loss path based on an identification method of a Bayesian network; outputting a structure weight matrix of the power loss, and identifying an inclination angle gt; a temperature gt; a set of influence factors of the wiring topology pair setting; calculating a power loss response index of each component based on the loss path weight; forming a response index thermodynamic diagram by using a space weight interpolation algorithm, and performing array space sensitive partitioning; the high response area is used as a preferential setting target, and the low response area enters an observation maintenance mechanism; by constructing the response index and the correction response potential function based on multi-factor coupling, the power response potential of the component to the setting operation can be accurately identified, the problems of average adjustment and low-efficiency resource allocation in a traditional setting method are effectively avoided, and it is ensured that setting resources are intensively input into a high-response area.

The invention discloses a remote sensing image pseudo label generation method based on prototype perception learning, and a weak supervision change detection model training and detection method, and belongs to the field of remote sensing image change detection. The invention aims to solve the problem that a category activation graph is incomplete in existing remote sensing image weak supervision change detection and the problem that an obtained detection model is unstable in change recognition. According to the method, feature maps of different scales before and after change are extracted from remote sensing images before and after change through a pseudo label generation model, change features are obtained, a class activation mapping map and a class activation mapping map of an unchanged class are obtained, and the class activation mapping map is segmented into response area masks through a threshold value and used for performing weighted aggregation on fusion features to obtain a class prototype; generating an enhanced class activation mapping graph based on the fusion feature and the class prototype, and selecting the class with the highest similarity as a pseudo tag of the pixel; and training based on the pseudo label data of the pixels to obtain a change detection model, wherein the model is used for remote sensing image change detection.

The invention relates to the technical field of learning methods, in particular to an OPGW heterogeneous detection method based on AI interference elimination, which comprises the following steps: acquiring a vibration signal segment, dividing mutation fragments, arranging channel sequence, comparing direction extension, extracting a direction change homing path segment, analyzing an offset trend positioning response area, and extracting fluctuation trend verification interruption. And combining state feature merging numbers, identifying a jump sequence and analyzing a switching relation, and obtaining disturbance boundary features. According to the method, a path chain is constructed through sudden change signal position division and a channel response sequence, a disturbance propagation starting point and an extension direction can be subjected to structured expression, a path segment corresponding to continuous direction change supports limitation of a response interval, a fluctuation trend and direction offset are combined to form segmented reference, and identification of characteristics of boundary evolution along with time is assisted; the hopping sequence in the tail end signal is used for restoring the disturbance boundary and enhancing the trajectory separation and feature retention capability of the multipath interference behavior in the differentiated environment.

The invention discloses a partition detection method for board card quality, and relates to the technical field of detection. The method comprises the following steps: dividing a board card into functional blocks according to a functional structure, and collecting thermal field distribution characteristic data of each block; constructing a standard thermal template, calculating the thermal imaging matching degree of the functional blocks by using an Euclidean distance algorithm, and marking the thermal imaging matching degree exceeding a preset threshold value as an abnormal thermal response region; collecting resistance and resistance gradient change data of an abnormal thermal response area, judging the area as a defect cluster area if conditions are met, and calculating a resistance abnormal amplitude; integrating defect cluster region data to form a composite characteristic spectrum, inputting a defect sample matching model, and outputting a defect classification result and a confidence score; calculating a regional quality score by combining fuzzy logic reasoning and a neural network algorithm and integrating a thermal imaging matching degree, a resistance abnormal amplitude and a confidence score; and fusing the defect classification result and the regional quality score to generate a quality grade label. And based on the quality grade labels, board card quality partition detection is completed.

The invention discloses an asphalt pavement maintenance strategy intelligent generation method and system, and relates to the technical field of road engineering.The maintenance strategy generation method comprises the steps that traffic analysis data information and traffic load data information are extracted from a multi-source traffic monitoring system, and the traffic analysis data information and the traffic load data information are obtained through a laser scanning and image intelligent recognition algorithm; analyzing the pavement structural disease level in each grid unit; according to the road surface structural disease level, the road condition risk level of each grid unit under the traffic load and damage superimposed effect is analyzed, a risk response area is divided, and a service life loss analysis instruction is generated; analyzing an accelerated decline trend of pavement structure deterioration of each grid unit in the risk response area to determine a life reduction value of each grid unit in the risk response area; after comparison, generating a pavement maintenance strategy of a corresponding grade; the maintenance strategy generation method has the advantages of being high in real-time performance, fine in area recognition and dynamically adjustable in strategy response.

The invention relates to an improved YOLOv8 small target detection method based on SENet, and the method comprises the steps: introducing semantic dilution loss, and measuring the dilution degree of small target features in a channel; a suppression reverse weight is generated through an SENet structure, and background redundancy is suppressed while a high response area is reserved; a C2f structure of YOLOv8 is fused, and decoupling characteristics are transmitted in cross-layer connection; in the Neck feature pyramid, a SENet response migration relation is constructed; calculating a weight deviation value, and judging whether the small target response has spatial deviation or not; position balance loss is introduced, and feature repositioning is carried out on a small target area with overlarge center-of-gravity drift; designing a channel response consistency measurement index, and measuring the SE response consistency of the small target between different epochs; sENet channel output is extracted from the image, and the variance of target area channel response distribution is counted; a channel with high variability is weakened or suppressed from a current image detection path, a detection head is introduced into a channel gating mechanism, a stable channel is adaptively selected to participate in prediction, and the small target feature representation capability is significantly enhanced.

The invention discloses an unmanned aerial vehicle competition decision-making method, device and equipment and a storage medium. The method comprises the following steps: constructing a competition situation evaluation model, a response region and an acquisition region of an unmanned aerial vehicle; based on the competitive situation evaluation model, the response area and the acquisition area, constructing an individual comprehensive reward model of the unmanned aerial vehicle; determining an unmanned aerial vehicle utility matrix based on the individual comprehensive reward model; based on a preset maneuvering strategy, a preset response strategy and the unmanned aerial vehicle utility matrix, constructing an unmanned aerial vehicle maneuvering-response decision dynamic game model; based on an improved grey wolf optimization algorithm, solving the maneuvering-response decision dynamic game model to obtain a maneuvering-response decision dynamic optimization result; and determining a target maneuvering-response strategy of the unmanned aerial vehicle based on a maneuvering-response decision dynamic optimization result.

The invention relates to the technical field of product life cycle management, in particular to an intelligent monitoring system and method for Liupu tea production, and the system comprises a disturbance sensing module, a layered structure module, an anomaly recognition module, a trend tracking module and a closed-loop linkage module. According to the method, airflow disturbance is introduced in a fermentation process stage, a direction hedging relationship is identified based on a temperature and humidity response difference value of nodes before and after the disturbance, an interference activation position is screened through a region hedging trend, three structural regions are divided by combining thickness, density and boundary response data, and a response lag region is identified to realize local focus point positioning. A structural reverse conduction area is screened out according to the direction change relation of collection nodes, an abnormal response area is recognized to achieve sudden change positioning, whether a spatial response mode changes suddenly or not is recognized through difference comparison between a disturbance result and a first response feature, and the perspectiveness and pertinence of abnormal response detection and the effectiveness of intervention effect evaluation are improved.

The invention relates to the technical field of circular screen equipment interaction, and discloses a circular screen audio playing interface optimization method and system, a storage medium and equipment, and the method comprises the steps: obtaining physical parameters of a circular screen, and building a polar coordinate system with the center of the screen as an original point; dividing a screen display area into a plurality of functional areas according to radial distances; based on the polar coordinate system and the partitioning result, carrying out adaptive layout on a plurality of elements of an audio playing interface; a touch response area of the interaction element is expanded, and multiple interaction gestures including a rotation gesture are recognized and responded; dynamic visual feedback is triggered in the user interaction process, and according to the method, the adaptation problem of a traditional rectangular UI on the circular screen device is comprehensively solved through the technical means of establishing a special coordinate system for the circular screen, optimizing the interface element layout, fusing multi-modal interaction gestures, enriching the visual animation effect, achieving dynamic self-adaptive optimization of the interface and the like.

The invention discloses a cascade operation state real-time monitoring and anti-collision risk intelligent early warning method and system, and the method comprises the steps: obtaining multi-sensor data, determining a key monitoring node, and extracting parameter changes to form an operation state trend sample; carrying out edge calculation processing on the trend sample, reversely analyzing and reconstructing component state parameters from the vibration spectrum signal, generating a step gap change curve, matching with a standard model to obtain a risk characteristic probability value, and generating a state traction chain according to the risk characteristic probability value; obtaining a displacement distribution coefficient and a vibration speed index from the state drag chain, predicting the collision probability distribution of the displacement distribution coefficient through cloud deep learning, determining an early warning level in combination with the vibration speed index, and generating early warning optimization configuration; emergency distribution is determined through linkage control analysis, differentiated response areas are identified, and linkage optimization factors are extracted from the differentiated response areas; finally, differential control parameters are generated and matched with the state traction chain to generate an early warning signal, and intelligent abnormal recognition and self-adaptive early warning of the step equipment are achieved.

The invention relates to the technical field of optical measurement of ultra-short-focus projectors, in particular to an optical measurement method and system for an ultra-short-focus projector, and the method comprises the following steps: obtaining the action time sequence of a main shaft and an auxiliary shaft, screening and starting adjacent combinations, analyzing the front-back relation to judge an interlaced sequence, comparing an angle track to recognize a reversal path, and comparing a direction trend. And screening axis numbers with consistent directions, and analyzing a brightness color change state to obtain measurement point change feature distribution. According to the method, motion combinations are screened through a multi-axis starting relation, posture adjustment nodes are judged by combining rotation direction changes, position changes are introduced into a measurement point response trend, an influence source is recognized through comparison of a direction trend and an offset direction, and differential response areas are marked out according to the stable state of brightness and color changes. The multi-parameter change forms association in a continuous path, omission caused by discrete observation in an information acquisition process is avoided, and the traceability of a dynamic change relation and quantitative coverage of parameter behaviors in an adjustment process are enhanced.

The embodiment of the invention provides an interaction method and device in a game, electronic equipment and a storage medium. A graphical user interface is provided through terminal equipment, and the method comprises the following steps: detecting an interactive event in a game scene, and determining a current touch position of a target interactive operation in the graphical user interface; generating a trigger response area according to the current touch position; and in response to a touch operation in the trigger response area, triggering an interaction operation corresponding to the interactive event. According to the technical means of dynamically generating the trigger response area, the user can directly trigger the interactive operation corresponding to the interactive event near the touch position without moving a finger to a fixed interactive button position on the screen, so that the large movement of the finger of the user on the screen is reduced, the problem of incoherent operation is avoided, and the user experience is improved. Meanwhile, the misoperation rate of the mobile terminal is reduced, and the response efficiency of game interaction is improved.

The invention provides a wheel center force extraction method, system and device for whole vehicle road noise simulation, and the method comprises the steps: building a vibration transmission relation between a wheel center excitation point and a steering knuckle response region based on a whole vehicle NVH finite element model, and determining frequency domain transmission data; collecting dynamic response signals of a plurality of steering knuckle response areas and road noise synchronization signals of a preset reference position in a time-sharing manner when the vehicle runs on an actual road surface, and performing phase alignment on frequency domain responses in different time periods according to the road noise synchronization signals to obtain a target frequency domain response matrix; determining a frequency domain excitation vector of the wheel center excitation point according to the frequency domain transmission data and the target frequency domain response matrix; and combining the frequency domain transmission data to reconstruct a theoretical response of a steering knuckle response region, and generating a target wheel center force. According to the method, the wheel center force suitable for high-frequency-band whole vehicle road noise simulation can be rapidly and accurately extracted with low cost on the premise of not depending on a high-cost sensor and a special test bench, so that optimization design of NVH (Noise Vibration and Harshness) performance of a new energy vehicle is supported.

The invention discloses a machine tool oil mist centralized management and control intelligent purification method and system, and relates to the technical field of oil mist management and control purification. State factors from different physical sources are subjected to normalization processing, the system breaks through a data dimension gap between sub-modules, and the management and control accuracy of the machine tool oil mist centralized management and control intelligent purification system is improved. The performance degradation behavior, the disturbance flow state behavior and the pollution residue behavior are uniformly incorporated into the same regulation and control intensity index system for the first time, and a uniform logic reference is provided for strategy evaluation. Different regulation and control intervals such as a normal area, a steady-state regulation and control area, a linkage control area and a warning response area correspond to different coping strategies respectively, abnormal data points caused by interference fluctuation, accidental jump and the like in oil mist parameter collection are effectively removed in a combined mode of a sliding window and median filtering, and the accuracy of oil mist parameter collection is improved. And it is ensured that data entering a subsequent calculation model has high stability and representativeness. The mechanism directly solves the common core problem of'high-noise data misleading judgment 'in the prior art.

The invention provides a nonlinear high-order vibration mode modulation method of an MEMS electric field sensor, and relates to the technical field of electric field sensors. The method comprises the following steps: constructing a three-dimensional simulation model of a resonant structure of the MEMS electric field sensor, and extracting vibration mode characteristics of a multi-order mode; according to the vibration mode characteristics of each order mode, the structure and the arrangement mode of a driving electrode and an induction electrode in the electric field sensor are optimized; applying an alternating current and direct current combined driving voltage matched with the target high-order mode to the driving electrode, so that the electric field sensor works in a Duffing nonlinear state of the target high-order mode; sensing signals are collected and processed, performance indexes including sensitivity, background noise and resolution are calculated, and the performance of the electric field sensor is evaluated. According to the invention, the electric field sensor is driven to work in a high-order mode and enters a Duffing nonlinear response area, so that high-sensitivity and high-resolution detection of an external electric field is realized, and noise is effectively suppressed.

The invention relates to the technical field of computer vision, in particular to a processing method and system based on a computer vision large model, and the method comprises the following steps: obtaining the alignment position of a continuous response region in a feature map, screening a consistent change region, analyzing semantic nodes, connecting a mapping path to an original map, extracting coordinate analysis direction grouping nodes, and tracking semantic change to mark a jump position, extracting feature analysis to activate a difference to delimit a mutation fragment, and obtaining a semantic response structure chart. According to the method, continuous regions with consistent directions are screened through position alignment of multilayer channel response regions, spatial focusing of semantic regions is carried out, path clues with continuous semantic relationships are screened through extraction of connection directions and sequence information of nodes in an output sequence, and path segments with a propelling trend are classified through analysis of node coordinate arrangement. And by tracking node semantic value changes, marking jump node positions, and dividing mutation fragments through response differences, the layering and distinguishing capability of the semantic structure is enhanced.

The invention discloses an underwater image enhancement method and system based on three-input multi-scale fusion. The method comprises the following steps: carrying out visibility recovery on an image to be processed; performing contrast enhancement through a nonlinear mapping function; carrying out enhancement processing on the contour of the high response area; respectively calculating a Laplacian contrast weight map, a saliency weight map and a saturation weight map of the three processed images; linearly combining all types of weight maps to obtain an aggregation weight map, and normalizing the aggregation weight to obtain a normalized weight map; decomposing the three processed images into a Laplacian pyramid, decomposing the normalized weight map into a Gaussian pyramid, and performing fusion through upward sampling and addition reconstruction; carrying out smoothing processing and edge-preserving sharpening processing on the fused image to obtain an image after underwater image enhancement and restoration; according to the method, the color cast problem of the underwater image is solved, the image contrast is effectively improved, and the texture details of the image are improved.

The invention discloses an intelligent simulation analysis method and system for standing wave characteristics and safety performance of an automobile tire, and relates to the technical field of intelligent simulation analysis and automobile safety. The intelligent simulation analysis method and system for the standing wave characteristic and the safety performance of the automobile tire comprises the following steps that S1, multi-dimensional simulation driving data are obtained and preprocessed; s2, reconstructing a tire structure response based on the preprocessed multi-dimensional simulation driving data, and measuring the intense degree of the structure response; s3, simulating tire temperature change in combination with a temperature rise simulation process, and judging whether temperature rise abnormity exists in the local area or not; and S4, carrying out space overlapping comparison on the structure response abnormity and temperature rise abnormity areas, and identifying a space overlapping state. The problems that a tire local high-temperature area and a standing wave high-response area are overlapped to cause structural instability easily, recognition is difficult to achieve in advance in the prior art, the monitoring mode is based on the overall trend, and a recognition and early warning mechanism for the local overlapping situation is lacked, so that the tire burst risk is difficult to perceive in time are solved.

The invention relates to the technical field of magnetic stimulation therapy, in particular to a magnetic convulsion therapy coulometric titration method and system based on individualized epilepsy prediction.Based on epilepsy prediction, intervention time periods are divided, electroencephalogram is collected to extract the phase change trend, and magnetic stimulation applying modes consistent in space direction are matched and screened; stimulation is executed, nerve electrical activity evolution is monitored, nerve response area distribution is concluded, a time period sequence, a response area form and an application trend are synthesized, a stimulation energy configuration range is divided and limited, and a magnetic convulsion treatment electricity titration result is formed. According to the method, epilepsy risks are described in a staged mode, time sequence correlation is established, intervention is endowed with dynamic distinguishing capacity, the stimulation direction is restrained according to phase propagation characteristics, the stimulation path is kept consistent with the dominant direction of neural activity, the energy range is limited in combination with neural response distribution, energy and state mapping is established, and the intervention matching degree and regulation controllability are improved; the invalid stimulation probability is reduced, and the treatment adaptation capability is enhanced.

The invention relates to the technical field of civil engineering monitoring and evaluation, and provides a comprehensive pipe gallery underground diaphragm wall external soil pressure data evaluation method and system, and the method comprises the steps: analyzing and recognizing a specific mechanical response region of a comprehensive pipe gallery underground diaphragm wall; aiming at a specific mechanical response area, establishing a mechanical correction rule; acquiring monitoring data of a soil pressure sensor, and calculating an initial soil pressure predicted value of a measuring point position corresponding to the monitoring data according to a classic soil pressure theory; according to the position of the measuring point, whether the measuring point is located in a specific mechanical response area or not is judged, the initial soil pressure predicted value is adjusted based on a mechanical correction rule, and a corrected soil pressure predicted value is obtained; and comparing the monitoring data with the corrected soil pressure predicted value to obtain a numerical comparison result, and judging whether the monitoring data is abnormal or not according to the numerical comparison result. The method has the effect of improving the safety of long-term operation of the comprehensive pipe gallery.

The invention discloses a method and a device for decoupling a slow-thinking area and a quick-response area in a large language model. The method comprises the following steps: constructing a slow thinking prompt and a fast response prompt, respectively inputting the slow thinking prompt and the fast response prompt into a large language model, and recording a neuron activation value of each linear layer of the large language model; calculating an activation value difference vector based on the neuron activation values corresponding to the slow thinking prompt and the fast response prompt, and obtaining a prompt difference vector based on the activation value difference vector; iteratively processing a plurality of reasoning problems, obtaining positive value position intersection of prompt difference vectors in all reasoning problems, and identifying a key neuron set related to slow thinking; and constructing a gradient mask matrix based on the position information of the key neurons in the key neuron set, and performing fine tuning on the key neurons by adopting a mask mode based on the gradient mask matrix to obtain an optimized large language model. The technical problem that an existing model is poor in reasoning ability is solved.

The invention relates to the technical field of disaster emergency management, and discloses a disaster emergency management method based on LSTM time sequence analysis. The method comprises the following steps: acquiring natural disaster situation data which is reported by a management node where a disaster situation is located and contains multi-dimensional time sequence characteristics; according to the multi-dimensional time sequence characteristics and a preset disaster situation type knowledge base, determining a target disaster situation type matched with the multi-dimensional time sequence characteristics; calling historical disaster situation data corresponding to the target disaster situation type, and performing time sequence analysis on the historical disaster situation data through an LSTM model to generate a disaster situation characteristic evolution sequence; and dividing a disaster response priority region and a secondary response region based on the disaster feature evolution sequence and the emergency measure database. According to the method, the disaster situation type can be accurately determined, the disaster situation evolution trend can be accurately analyzed, the response area can be scientifically divided, the emergency resource distribution can be optimized, the disaster situation emergency management efficiency can be improved, and the method is suitable for emergency management work of various natural disaster situations.

The invention provides an artificial intelligence confrontation technology military training simulation system based on reinforcement learning, and belongs to the field of training simulation. The problem of poor confrontation simulation training effect is solved; the scene simulation module is used for constructing a differential equation set and setting a response area of a target or a response interval of a laser module; the training simulation module is used for acquiring the positions of trainees, counting, recording and analyzing the shooting interval and the effective score of each trainee in combination with a differential equation set and the response area of the target or the response interval of the laser module, and adjusting a training strategy; the score feedback module is used for recording and feeding back effective scores of the trainees; the target refreshing interval and position are adjusted by analyzing the score and shooting angle of the trainee, and the target moving route is distributed according to the position of the trainee, so that the model effect of confrontation training is improved.

The invention relates to the technical field of computer vision, and discloses a method and system for adjusting a video picture display proportion of a splicing window, and the method comprises the steps: recognizing the original resolution and content features of a to-be-spliced video stream; establishing a feature matrix for receiving the to-be-spliced video stream, and establishing a layout analysis network of the to-be-spliced video stream, so as to determine an initial layout of the to-be-spliced video stream through the feature matrix; calculating an adaptive proportion interval of the to-be-spliced video stream; constructing a Grad-CAM thermodynamic diagram of the to-be-spliced video stream, marking a high response region of the Grad-CAM thermodynamic diagram, and performing nonlinear scaling on the to-be-spliced video stream according to the high response region and the adaptive proportion interval to obtain a scaled to-be-spliced video stream; and calculating an attention index for zooming the to-be-stitched video stream to generate a proportion adjustment instruction and a transition animation parameter for zooming the to-be-stitched video stream, and executing picture display proportion adjustment for zooming the to-be-stitched video stream based on the proportion adjustment instruction and the transition animation parameter. The display effect of the video picture can be improved.

The utility model discloses a small boxing target. The small boxing target comprises a first surface pasting layer and a second surface pasting layer, the striking area is assembled on one surface of the first surface pasting layer; a second facing layer; the response area is assembled on one surface of the second surface pasting layer, and the striking area is in contact connection with the response area; wherein one side of the first surface pasting layer and / or one side of the second surface pasting layer are / is provided with at least one circle of first bonding layer which is used for bonding and fixing the first surface pasting layer and the second surface pasting layer. By arranging the protrusions, use of an interlayer is avoided, the hitting hand feeling is improved, strength feedback distortion is reduced, materials can be saved, and the cost can be reduced.

The utility model belongs to the technical field of touch sensing, and discloses a distance touch sensing device and a household appliance, which comprise a touch panel, a sensing probe and a sensing identification module, the touch panel is provided with a plurality of touch sensing areas, the touch sensing areas are used for receiving external touch input operation, at least one touch sensing area corresponds to a sensing probe, and the touch sensing areas correspond to at least one sensing probe; one end of each sensing probe is connected with the circuit board, the other end of each sensing probe extends in the direction far away from the circuit board and close to the touch panel and is connected with the touch panel in the corresponding touch sensing area by taking air as a conducting medium, and the sensing probes are used for sensing touch input operation and generating corresponding sensing signals; the induction recognition module is arranged on the circuit board, the induction probe is connected with the induction recognition module through the circuit board, and the induction recognition module is used for receiving the induction signal and determining touch input operation according to the induction signal. The utility model aims to improve the accuracy of touch sensing.

The invention discloses a visual segmentation large model data automatic annotation method combined with image prompt, which comprises the following steps: extracting deep features of a prompt image and a to-be-annotated image by adopting a shared visual coding technology of deep learning, and carrying out cross-image feature matching and similarity calculation. And positioning a high-response area which is consistent with a prompt target in semantics in the to-be-labeled image, and automatically screening out an accurate prompt point set from the high-response area. Furthermore, based on the point set, the visual segmentation large model is guided to complete accurate segmentation and mask generation of the target, so that full-process automation from image prompting to final labeling is realized, and labeling efficiency and consistency are remarkably improved. In this way, the bottleneck problems that in the prior art, an automatic labeling method is insufficient in universality and the application of a segmentation large model must depend on human interaction are solved, and an effective technical approach is provided for large-scale, low-cost and high-quality visual segmentation data production.