8results about How to "Eliminate mismatch" patented technology

The application relates to the technical field of image detection, and discloses a scale-invariant contour matching method and device based on machine vision and a storage medium, the method comprising the following steps: extracting a target contour from a template and a query image and detecting key points; constructing a multi-scale contour context descriptor for each key point, the descriptor being composed of chord length ratios and chord angle features under multi-scale in a local neighborhood; performing feature matching, estimating scale and rotation transformation parameters according to chord length ratios and chord angles of a matched descriptor pair, and combining key point coordinates to calculate a translation parameter to generate a similarity transformation hypothesis; clustering all hypotheses as points in a high-dimensional parameter space to obtain candidate instances; for each candidate instance, fitting an accurate similarity transformation model and eliminating false matching by using a random sample consensus algorithm through matched point pairs associated with the candidate instance, and outputting final positioning parameters of a target object. The application realizes efficient, accurate and scale- and rotation-invariant contour matching.

This invention relates to the field of industrial visual inspection technology and discloses a visual detection method for particulate contaminants in coolant. The method includes: acquiring parallel polarization image sequences and orthogonal polarization image sequences by performing dual-channel synchronous polarization image acquisition using a beam splitter; performing inter-frame cross-power spectrum subpixel displacement estimation on the parallel polarization image sequences to generate a frame-by-frame subpixel motion field vector sequence; performing subpixel-level translation correction on the dual-channel image sequences based on the subpixel motion field vector sequence to generate a background-stabilized dual-channel polarization sequence; performing pixel-by-pixel temporal fluctuation intensity statistics and short-time frequency analysis to generate a set of particulate contaminant events; extracting dual-channel grayscale time-series signals and calculating polarization degree time-series signals; performing phase-compensated coherent superposition to generate a cumulative polarization degree sequence; extracting multidimensional optical polarization feature vectors and inputting them into a pre-trained classifier to output material category labels and classification confidence scores; and generating a detection report.

This invention discloses a fine motor imagery decoding system based on PINN and multimodal fusion. The system employs a fine motor imagery decoding method based on PINN and multimodal fusion, and includes a multimodal physiological signal synchronous acquisition module, a data preprocessing and correlation analysis module, a dual-stream decoding network module, a physical manifold constraint layer module, an adaptive gradient balance optimization module, and a classification decision module. By deeply embedding biophysical mechanisms into a deep learning framework, this invention addresses the ill-posedness of the source localization inverse problem, overcomes overfitting in small-sample scenarios, and significantly improves the decoding accuracy, robustness, and physiological interpretability of fine hand movements.

The application relates to the technical field of intelligent detection, and discloses a system and a method for adjusting a recognition algorithm after intelligently detecting and recognizing a scene, wherein the method comprises the following steps: acquiring current frame detection data and extracting multi-dimensional scene feature descriptors, and maintaining a scene descriptor time sequence cache queue; performing time sequence difference analysis on the queue to generate a scene mutation prediction signal; based on the signal state, pre-fetching a target scene recognition algorithm parameter set to a pipeline preparation register; when a new frame arrives, promoting the parameter set in the preparation register to an effective parameter and performing feature separation operation; performing multi-dimensional target recognition on each feature response separation graph and calculating a weighted fusion confidence score; screening a candidate target region and outputting a detection and recognition result.

The invention provides a mechanical arm trajectory tracking control method fusing model prediction and sliding mode control. The mechanical arm trajectory tracking control method comprises the steps that a discrete sliding mode controller is constructed, and a discrete sliding mode surface based on trajectory tracking errors is designed; calculating a sliding mode control law; constructing a dynamics prediction model based on an echo state network, and performing online learning and updating by taking historical state data and a sliding mode control law of the mechanical arm as input so as to predict a state track of the mechanical arm in a future time domain; constructing a model prediction controller embedded with sliding mode control, and introducing a prediction state and a sliding mode control law into an optimization objective function of the model prediction controller; and under the model prediction controller, an optimization problem is converted into a quadratic programming problem to be solved, a smooth optimization control quantity meeting physical constraints is obtained, and the control quantity acts on the mechanical arm system. The high-frequency buffeting problem of sliding mode control is effectively solved, and meanwhile high-precision trajectory tracking of the mechanical arm under strong nonlinear interference is guaranteed.

The application relates to the technical field of semiconductor devices, in particular to a method for preparing a gallium oxide material for a power device based on phase conversion and application thereof. The method comprises the following steps: depositing a polycrystalline Ga2O3 film on a substrate and heavily doping the same as a substrate material for subsequent device preparation; bonding a light-doped beta-Ga2O3 film with a preset thickness on the polycrystalline Ga2O3 film through bonding, and using the lattice arrangement of the beta-Ga2O3 as a traction for the subsequent polycrystalline Ga2O3 phase conversion process; converting the polycrystalline Ga2O3 into beta-Ga2O3 under preset conditions, and repairing the damage of the film interface during bonding; and removing the substrate, thereby completing the material preparation of the device. The method utilizes the property that the metastable phase of gallium oxide isomers can be converted into beta-phase isomers under certain conditions, converts the polycrystalline Ga2O3 film into a high-quality beta-Ga2O3 substrate, eliminates the lattice mismatch between different phases, and finally obtains a low-cost high-quality beta-Ga2O3 film which can realize a large power device.

The application discloses a kind of compressed computing spectral imaging reconstruction method and system based on pulse weighted network, it is related to computing imaging technical field, it can be from compressed measurement value with low energy consumption and high quality Reconstruct HSI data, can greatly improve the energy efficiency ratio of reconstruction algorithm.The scheme is specifically as follows: using coded aperture snapshot spectral imager CASSI obtains the two-dimensional compressed measurement value of known target scene three-dimensional hyperspectral imaging technology HIS data cube.The two-dimensional compressed measurement value obtained is preprocessed, image features are extracted, and the initial value of the reconstruction result of the HSI cube of the known target scene is obtained.The initial value of the reconstruction result is spliced with the coded pattern of the compressed spectral imaging system, and is sent into the trained multi-time-step reconstruction network to obtain the output result of the network, i.e., the reconstruction result of the hyperspectral data of the target scene to be measured.

The application relates to a multi-step tool calling data processing method and system based on a large language model, and relates to the fields of artificial intelligence and data processing, which comprises the following steps: in response to a calling instruction of a first tool, the first tool is executed to obtain a first complete result; the first complete result is written into a runtime storage area independent of a context of the large language model, and a corresponding first reference identifier is generated; a result abstract of the first complete result and the first reference identifier are returned to the large language model, so that the large language model generates a calling parameter for a second tool based on the result abstract; and in response to the first reference identifier contained in the calling parameter, the second tool reads the first complete result from the runtime storage area through the first reference identifier. The application has the effects of realizing decoupling transmission of intermediate results and model contexts and deterministic conversion of data structures in multi-step tool calling, reducing resource consumption and improving execution stability.