27results about How to "Overcome precision" patented technology

The application relates to a basketball video highlight automatic generation method and system based on multi-modal feature fusion, and belongs to the technical field of computer vision and multimedia intelligent analysis. The method comprises the following steps: S1, extracting and aligning a visual feature sequence and an audio feature sequence; S2, using bidirectional LSTM to perform time sequence coding on the audio-visual feature, constructing a cross-modal attention mechanism taking audio as a query and vision as a key value, and obtaining multi-modal fusion features; S3, inputting the fusion features into a 1D CNN to output a frame-by-frame highlight event probability sequence; S4, performing peak value searching based on an adaptive threshold of the probability sequence to obtain a coarse-grained timestamp; S5, taking the coarse-grained timestamp as a center to intercept local features, inputting the local features into bidirectional LSTM to predict starting and ending offsets, and obtaining accurate start and end time boundaries; and S6, clipping and splicing to generate a highlight video according to the accurate boundaries. The application realizes audio-visual semantic depth alignment through an audio-guided visual attention mechanism, and improves boundary positioning accuracy through a coarse-precision two-stage positioning architecture.

This invention discloses a quantitative test strip for HBV PreS1 antibody based on the AIE molecule, its preparation method, and its applications, relating to the fields of molecular biology, immunology, and biodetection technology. The test strip includes a sample pad, a labeling pad, a nitrocellulose membrane (NC membrane), an absorbent pad, and a base plate. The NC membrane detection line is coated with HBV PreS1 antigen to capture PreS1 antibodies in the sample. The labeling pad contains an immobilized Anti-Human IgG antibody conjugated to the AIE molecule, which binds to the PreS1 antibodies in the sample to form an immune complex. The test strip of this invention enables rapid detection of PreS1 antibodies and achieves enhanced fluorescence in the detection results. Combined with a fluorescence reader or fluorescence card reader, the detection results can be digitally and quantitatively analyzed. It is simple to operate, rapid in detection, and significantly improves detection sensitivity and enables quantitative analysis.

The invention discloses a refrigerant adjusting method, device and system. The method comprises the steps that the current refrigerant temperature difference between the first refrigerant temperature of an inlet of the air-cooled condenser and the second refrigerant temperature of an outlet of the air-cooled condenser is obtained; the current deviation between the current refrigerant temperature difference and the temperature difference reference value is determined; under the condition that the current deviation is not within the first range, a feedback control model based on a real-time error is called, the opening degree of a first adjusting valve is adjusted according to the current deviation, so that the deviation between the adjusted refrigerant temperature difference between the inlet and the outlet of the air-cooled condenser and the temperature difference reference value is within the first range, and the first adjusting valve communicates with the air-cooled condenser; according to the scheme, the fluctuation of the key heat exchange parameter (temperature difference) is actively restrained within a set reasonable range, system performance degradation or operation risks caused by large drifting of parameters are effectively avoided, stable and reliable operation of the whole system is guaranteed, and the technical problem that the operation stability of an evaporation cold and heat pump unit system is low can be solved.

This invention belongs to the interdisciplinary field of artificial intelligence and geological resource assessment, specifically relating to a method and system for predicting the energy storage potential of salt caverns based on multi-source geological information. It aims to address the problems of low prediction accuracy and difficulty in balancing breadth and precision in existing technologies due to single-source data, static models, and lack of adaptive capabilities. The method includes: acquiring multi-source data such as seismic, well logging, core, and geostress data; constructing a geological feature fusion vector containing seven core parameters; establishing a high-dimensional nonlinear mapping model driven by a deep neural network; dynamically adjusting the prediction range based on data coverage density and confidence level; and achieving continuous model iteration through online incremental learning. The system integrates data quality assessment, anomaly handling, and distributed computing modules. By adopting the above technical solutions, this application can achieve a significant improvement in prediction accuracy and adaptive intelligent adjustment of the prediction range.

The invention relates to the technical field of semiconductor detection, discloses a semiconductor image processing method, and particularly relates to an image template matching result evaluation method and system and a readable storage medium. According to the embodiment of the invention, the edge point corresponding to each edge point in the to-be-matched image, namely the edge image, in the template image is constructed, the inner point set of the template image is constructed based on the edge points to serve as the target feature for evaluating the to-be-matched image, and then the point which changes most violently in the inner point set is determined as the optimal edge point. The optimal edge point is used as an optimal point for evaluating the positions of the template image and the image to be matched; based on the edge point position error of the optimal edge point and the matched edge point, the deviation between single point positions is determined, finally, the average deviation of the deviation of the multiple point positions serves as the final deviation between the two images, and evaluation of the accuracy of the matching result is achieved through the deviation.

The utility model discloses an alignment mechanism for pole piece detection, include: base plate, and its both ends end face respectively fixedly be equipped with mounting seat, the first slider and the second slider of sliding setting in the mounting seat, the adjusting end of mounting seat is fixedly equipped with first drive assembly, and the output shaft of first drive assembly is connected and drives corresponding slider sliding, the first pivot assembly of fixedly equipped on the first slider and two second pivot assemblies of fixedly equipped on the second slider, be equipped with the alignment plate of center round hole and racecourse type through -hole, first pivot assembly is worn in center round hole, and two second pivot assemblies are respectively worn in the racecourse type through -hole that sets up parallel to the alignment plate length direction, the pole piece detection camera of fixedly equipped in the alignment plate middle part edge position, through the automatic control of first drive assembly and second drive assembly, has realized the automation of alignment process, has improved alignment efficiency and precision greatly.

This invention belongs to the field of equipment inspection technology and discloses an intelligent inspection and positioning method and system based on Bluetooth and NFC. The method includes: receiving Bluetooth signals from available Bluetooth devices through an intelligent inspection and positioning terminal, and sorting valid inspection points based on the access time of each Bluetooth signal to generate a time series representing the inspector's movement route, thereby automatically tracking the inspector's movement route; when at least two assigned inspection areas are identified, comparing the number of shared inspection points between different assigned inspection areas with an intersection threshold value to determine key inspection areas or ordinary inspection areas, and then generating an inspection route. This invention reconstructs the inspector's movement route based on the time series of Bluetooth signal access, improving the authenticity and objectivity of the inspector's movement route, and by transforming inspection planning into a dynamic response, improving the overall efficiency and targeting of inspection work.

The invention discloses a compound control method and system for a mechanical arm with a gap joint, and belongs to the technical field of engineering equipment control. Based on a six-degree-of-freedom mechanical arm nonlinear dynamic model containing a gap joint, control boundaries of a collision force feedforward module and a PD feedback module are determined by quantifying a coupling relation between a joint gap and collision force; then a composite control structure of'collision force feedforward compensation-PD feedback tracking 'is built based on the kinetic model, a feedforward module counteracts disturbance of nonlinear collision force on motion of the mechanical arm in advance by collecting joint collision force signals in real time and calculating reverse compensation torque, and a PD feedback module counteracts disturbance of nonlinear collision force on motion of the mechanical arm in advance by monitoring joint positions and posture deviation in real time; outputting a tracking torque to accurately correct a track error; the two layers of modules are nested and cooperatively work, the advantage of active disturbance suppression of feedforward control is exerted, the steady-state precision of trajectory tracking is guaranteed by means of PD feedback control, and the operation precision and reliability of the gap-containing joint mechanical arm are effectively improved.

The invention provides an L1 / L2 norm ratio regular minimization image restoration method based on an analytical model, and belongs to the field of signal and image processing inverse problems. In order to solve the problems that the reconstruction precision is limited and the algorithm robustness is insufficient under the conditions of undersampling, pixel loss, blurring and the like in the existing image restoration technology, an image restoration optimization model is established and solved by utilizing the inherent scale invariance and parameter freedom of an L1 / L2 norm ratio and combining the stability of signal representation under an analytical model. The geometric structure and texture details of the image can be stably recovered from under-sampled or damaged observation data, and the image recovery problem in various typical degradation scenes is effectively solved.

The invention belongs to the technical field of power plant indirect cooling circulating water treatment, and relates to an automatic control device and method for the quality of indirect cooling circulating water. Comprising an intelligent control cabinet, an alkali solution tank, an acid solution tank, a mixed ion exchanger, a condenser and an indirect cooling tower aluminum heat exchanger, a conductivity meter, a pH meter and a PLC (Programmable Logic Controller) are arranged in the intelligent control cabinet; the alkali solution tank is connected with a circulating water inlet pipeline; the acid solution tank is connected with a circulating water inlet pipeline; a pH electrode is arranged on the circulating water inlet pipeline; a conductivity electrode is arranged on the circulating water outlet pipeline; the condenser inlet is connected with the circulating water outlet pipeline; the condenser outlet is connected with the circulating water inlet pipeline; the circulating water outlet pipeline is connected with an inlet of the mixed ion exchanger; an outlet of the mixed ion exchanger is connected with an inlet of the condenser. According to the invention, accurate control of the water quality pH value and conductivity of the intercooling circulating water can be realized, so that system corrosion is effectively inhibited, the service life of equipment is prolonged, and safe and economical operation of a unit is guaranteed.

The application discloses a kind of power grid architecture data processing method and system, belong to data processing field, this method includes: obtaining current project information, and current query set is constructed based on current project information;Obtain historical knowledge graph, extract structure information and construct historical query set;According to semantic query set obtained by large language model;Current query set, historical query set and semantic query set are confidence score, and initial decision group is obtained and fitness function is constructed;Obtain initial evaluation particle swarm, based on fitness function and compression factor iterative calculation, when reaching maximum iteration number or the average fitness of initial evaluation particle swarm meets preset fitness requirement, final evaluation particle swarm is obtained, to obtain final decision group, to complete power grid architecture data processing process based on final decision group.The power grid architecture data processing method disclosed in the application can optimize the data processing architecture of power grid, and improve data processing efficiency.

The image motion deblurring method based on a hybrid expert network belongs to the technical field of image processing and computer vision, and particularly relates to an image motion deblurring method based on a hybrid expert network; solves the technical problems of coarse-grained routing mechanism, sparse activation detail loss, imbalance between lightweight and precision, single feature fusion strategy, and difficulty in balancing precision and efficiency in the prior art; the method comprises the following steps: inputting a single-frame blurred image to be repaired into a hybrid expert deblurring model that has been trained, and reconstructing and outputting a clear image after deblurring. The image motion deblurring method based on a hybrid expert network is suitable for the fields of lightweight image visual perception, image restoration and deblurring processing.

The embodiment of the invention provides a robot eyeball tracking control method and device, computer equipment, a computer readable storage medium and a computer program product, and belongs to the technical field of robots. The robot eyeball tracking control method comprises the following steps: determining a three-dimensional pose of a target object in a current camera coordinate system by using a binocular camera, and determining an expected camera extrinsic parameter of the target object at a target position based on the three-dimensional pose; obtaining current steering engine control data of eyeballs of the robot, and determining current camera external parameters from a pre-constructed mapping relation table based on the current steering engine control data; and determining the difference between the expected camera external parameter and the current camera external parameter, determining a steering engine control instruction of the robot eyeball at the next moment based on the difference and the mapping relation table, and driving the robot eyeball to move based on the steering engine control instruction. According to the technical scheme, the control problem caused by mechanical uncertainty of the composite steering engine can be effectively solved, and high-precision tracking control over the object by the eyeballs of the robot is achieved.

The application discloses a preparation method of a silicon carbide membrane based on wettability difference induced hollow interface construction. The method introduces a carbon powder pore former with wettability difference from silicon carbide particles into a green support body, realizes slurry preferential spreading in the hydrophilic region and local retention in the hydrophobic region during the dip-coating process, and then induces the formation of an ordered hollow channel structure at the membrane-support interface. During the co-sintering process, the structure can effectively inhibit the penetration of the membrane slurry into the large pores of the support body, enhance the integrity of the membrane layer, and significantly reduce the interface mass transfer resistance. The obtained membrane layer has adjustable pore size and highly continuous surface structure, significantly improving the gas flux and dust filtration performance compared with traditional membranes, while shortening the preparation period and reducing the sintering energy consumption. The method is suitable for the construction of various asymmetric membrane structures, has good scalability and process compatibility, and provides a new way for the large-scale preparation of low-cost and high-performance silicon carbide membranes.

The application discloses a forming method of a large-size high-temperature alloy curved generatrix single-wall section product, which comprises the following steps: preparing a fan-shaped spread material; rolling and bending the fan-shaped spread material to form a conical cylinder by forming and combining welding; performing flanging forming on the conical cylinder, and the flanging angle is 30-35 degrees; performing bulging forming on the product obtained by the flanging forming; and performing solid solution treatment on the product obtained by the bulging forming, and the shape correction is realized by adopting the method of re-bulging forming, so that the springback is reduced. The application further discloses a forming device of a large-size curved generatrix single-wall section product, which comprises a flanging die, a female die and a bulging die; wherein the flanging die and the female die are used for realizing flanging forming; and the bulging die and the female die are used for realizing bulging forming. The product obtained by the application has high shape and size precision, low forming period and manufacturing cost, and small springback of the high-temperature alloy product after forming, and is suitable for batch production of large-size high-temperature alloy single-wall section products.

This application relates to the field of medical image processing technology, and discloses a method, system, device, and medium for predicting tumor status, including: acquiring a set of CT images of the lungs of a target object during the current respiratory cycle; constructing a three-dimensional tumor model of the lung tumor based on the CT image set; the lung tumor has a real volume and real centroid coordinates; performing finite element analysis to obtain a finite element tumor model, calculating the simulated volume and simulated centroid coordinates of the lung tumor; comparing and obtaining the volume retention ratio and centroid offset; predicting the motion state of the lung tumor in the next respiratory cycle to obtain the initial motion state parameters of the lung tumor in the next respiratory cycle; if the motion state prediction quality of the lung tumor meets the preset motion state prediction quality based on the volume retention ratio and centroid offset, then the initial motion state parameters are used as the target motion state parameters to predict the motion state of the lung tumor, thereby improving the accuracy and correctness of radiotherapy.

This invention belongs to the field of microwave radar image processing technology, and relates to a method for inverting significant wave height based on wave period. First, a significant wave height calculation model is established with the wave period as input. Then, microwave remote sensing images of the sea surface are acquired and preprocessed, such as coordinate transformation (e.g., images from shore-based sea observation radar and shipborne navigation radar) and noise suppression. Next, the wave period is inverted from the preprocessed images using spectral analysis techniques. Finally, the inverted wave period is substituted into the pre-established model to calculate the significant wave height. This invention, by introducing the wave period as a key intermediate variable, closely correlates image features with ocean wave physical parameters, overcoming the shortcomings of insufficient accuracy and stability in existing direct inversion methods. It provides a physically meaningful, highly accurate, and easily operational method for obtaining significant wave height.

This application discloses a rapid storm surge floodplain ensemble forecasting method and system for operational use, comprising: generating an ensemble typhoon path containing multiple deviation paths based on real-time typhoon forecast paths and historical error information; calculating the model wind field for each path; using a GPU-accelerated storm surge model to drive parallel calculation of hourly storm surge rise fields and representative tide gauge station rise sequences under each path and determine their maximum total tide level; simultaneously calculating the minimum geometric distance between each pre-divided cell and all typhoon paths; selecting target floodplain cells based on three indicators: maximum storm surge rise, maximum total tide level, and minimum typhoon distance; extracting open-boundary driving data of the target cells from the regional rise field to calculate the storm surge floodplain results for each target cell; and finally integrating the results of all target cells to output the final floodplain calculation result. This method achieves a high-precision and high-efficiency operational rapid forecast.

The invention provides a finished bridge linear smoothness judgment method and system based on a main beam splicing process, and relates to the technical field of linear smoothness judgment, and the method comprises the following steps: obtaining a design curve of a main beam finished bridge target elevation, an elevation curve of an actual main beam splicing section and a to-be-judged splicing section length; and if the actual elevation curve exceeds the preset threshold, determining the elevation curve of the first bridge section according to the design curve. And fitting the actual elevation curve with the elevation curve of the first bridge section by adopting a preset spline curve fitting method to generate an elevation curve of the second bridge section. And combining the elevation curves of the actual splicing section, the first bridge section and the second bridge section into a virtual bridge forming line shape, performing high-pass filtering on the line shape, extracting an irregularity maximum absolute amplitude, comparing the irregularity maximum absolute amplitude with a preset irregularity allowable limit value, and judging whether the smoothness of the splicing section meets the requirement or not. According to the method, the smoothness judgment precision of the finished bridge line shape can be effectively improved.

The invention discloses an oblique light spot energy distribution evaluation system for additive manufacturing, and relates to the field of laser process measurement, and the system comprises a motion execution module which is used for driving a laser emission unit to carry out pose adjustment in a three-dimensional space; the detection judgment module is used for judging whether the spot size exceeds a preset measurement range or not in real time in the laser scanning process and outputting a mode selection signal according to the spot size; a dual-mode measurement module selectively enabling a first measurement subunit or a second measurement subunit in response to the mode selection signal; and the data processing and modeling module is used for establishing a function relationship between the light spot size and the energy attenuation along with the change of the defocus distance, and calculating an energy distribution analytical expression of the laser on the target working plane when the laser is incident at any inclination angle. According to the scheme, the process of evaluating the energy distribution of the oblique light spots can be innovated into rapid analytical calculation based on actual measurement, so that a key basis is provided for real-time optimization of additive manufacturing process parameters.

The invention discloses a glass liquid level non-contact measurement system and method, and belongs to the technical field of substrate glass manufacturing process control, and the system comprises a laser triangular reflection module, an infrared thermal imaging module, a data fusion processor and an air curtain protection device. A multi-mode non-contact measurement system is constructed, high-precision stable measurement with the liquid level of + / -0.1 mm is achieved in the complex environment of a kiln channel, the positive significance is achieved for improving the automation level and product competitiveness of a high-end glass manufacturing process, and the problems that in the high-temperature environment, a platinum probe is prone to deformation and corrosion, and the product quality is poor are solved. The problems of measurement errors caused by the fact that glass adheres to the surface of a probe to form insufficient fusion and need to be frequently maintained, contact measurement interferes with melt flow stability, and an existing non-contact technology (such as ultrasonic waves) is insufficient in precision under high-temperature and high-dust working conditions are solved.

The application discloses a multi-layer multi-scale grid and a dynamic auxiliary positioning method based on the same, and belongs to the field of wireless positioning. The method comprises the following steps: performing grid division on a target GNSS reference station network coverage area to obtain longitude difference dl, latitude difference db and height difference dh of the target GNSS reference station network coverage area grid; determining a multi-layer multi-scale grid of the target GNSS reference station network coverage area; based on the multi-layer multi-scale grid, a data center receives rough position information sent by a user, searches for a grid point closest to the user as a positioning service grid, and sends information of a virtual observation station of the positioning service grid to the user, so that real-time positioning is completed. The application solves the centimeter-level positioning demand of the service end for a large number of users from the principle of VRS technology, overcomes the problems of low terminal RTK fixing rate and poor accuracy in a complex terrain area, reduces the calculation pressure of the service end, has simple algorithm and strong practicability, and effectively reduces the construction and operation and maintenance cost caused by simply relying on the accumulation of server and other hardware facilities.

The invention relates to the technical field of Internet of Things intelligent gas station network energy management and regional supply and demand balance analysis, in particular to a regional supply and demand balance analysis device and method supporting energy double-control, and the method comprises the steps: a protocol self-adaption module solves the problem of gas station equipment protocol heterogeneity through dynamically selecting a communication protocol, and outputs a data stream in a unified format; the data sensing module adaptively adjusts the acquisition frequency according to the supply-demand imbalance risk index; the transmission optimization module adopts a multi-link transmission path selection mechanism to guarantee data transmission integrity; the cleaning reconstruction module improves the data quality through anomaly detection and credibility weight calculation; the prediction simulation module realizes accurate demand prediction based on weighted reconstruction data and a digital twin environment; and the strategy generation module synthesizes the multi-source information to generate an optimization strategy. All the modules form closed-loop control through bidirectional data streams, and the regional fuel supply and demand prediction precision and the energy double-control strategy timeliness are effectively improved.

The invention provides a railway bridge construction operation and maintenance method and system based on carbon emission optimization, relates to the technical field of traffic infrastructures, and aims to construct a bridge digital twin model in a construction stage to simulate and predict carbon emission dynamics. And carrying out correlation analysis on the carbon emission influence factors by using a space-time diagram network. The carbon emission is globally optimized through a quantum annealing algorithm, construction scheduling is optimized in combination with a quantum decision tree, dynamic regulation and control of the carbon emission are achieved, and meanwhile the influence on bridge material attenuation is determined. And an operation and maintenance stage: inputting the bridge structure attenuation information and the bridge structure data obtained in the operation and maintenance stage into a convolutional neural network, predicting a bridge structure state, and generating a dynamic maintenance decision according to the bridge structure state. The method can accurately monitor and optimize the carbon emission in the construction, operation and maintenance stages of the railway bridge, and provides technical support and decision basis for the industry to realize low-carbon and sustainable development targets.

This invention belongs to the field of underwater acoustic target tracking technology, specifically relating to a multi-target tracking method, program, device, and storage medium based on velocity measurement and random finite set theory. This invention constructs a joint probability density model of Poisson-non-Poisson hybrid clutter and combines it with an adaptive clutter rate estimation algorithm to achieve real-time online estimation of clutter intensity, distribution parameters, and clutter rate, overcoming the problems of clutter misjudgment and decreased tracking accuracy caused by model mismatch in traditional methods. This invention constructs a multi-target joint filtering framework that deeply integrates CPHD and GLMB, balancing target cardinality estimation accuracy with track management capabilities. It synchronously transmits multi-target probability hypothesis density, cardinality distribution, and tagged track identifiers throughout the prediction and update process, improving the stability of target quantity estimation and achieving full lifecycle management of single-target tracks, effectively reducing track fragmentation, ID exchange, and false alarm rates under dense clutter.

The invention relates to an additive manufacturing process optimization method based on heterogeneous interface thermal stress analysis, and belongs to the technical field of additive manufacturing processes. Comprising the following steps: constructing a three-dimensional geometric model of a substrate and a printing layer in Abaqus; performing grid division on the geometric model, and calculating physical property parameters of a printing layer material; compiling a heat source model, selecting different process parameters, and performing temperature field analysis on the model; on the basis of temperature field distribution, a sequential thermal coupling model is adopted, and the thermal stress of a heterogeneous interface in the additive manufacturing process is calculated; and analyzing normalized equivalent residual stress, distributed along the laser scanning direction, of a substrate on the lower side of a model interface and a printing layer on the upper side of the model interface in a stress field result and normalized shear stress, distributed along the laser scanning direction, of the interface to obtain optimal process parameters. According to the method, the finite element model is established, and a new method is provided for optimization of the TC4 / CrCuV heterogeneous material additive manufacturing process.

The application discloses an AIE molecule-based HBV PreS1 antibody quantitative detection test strip as well as a preparation method and application thereof, and relates to the fields of molecular biology, immunology and biological detection technology.The test strip comprises a sample pad, a mark pad, a nitrocellulose membrane (NC membrane), a water absorption pad and a bottom plate, the NC membrane detection line is coated with an HBV PreS1 antigen and is used for capturing PreS1 antibodies in a sample, and the mark pad is fixed with Anti-Human IgG antibodies coupled with AIE molecules and is used for combining with the PreS1 antibodies in the sample to form an immune complex.The test strip can realize rapid detection of the PreS1 antibodies, realize a fluorescence enhancement effect of a detection result, and, in combination with a fluorescence reading instrument or a fluorescence reading card equipment, the detection result can be subjected to digital quantitative analysis, is simple and convenient to operate, rapid to detect, significantly improves detection sensitivity and realizes quantitative analysis.