92 results about "Error map" patented technology

The present invention relates to the technical field of image processing, and in particular to a hyperspectral remote sensing image anomaly detection method based on deep learning. The method comprises: performing normalization preprocessing on a hyperspectral remote sensing image, so as to obtain an original hyperspectral image; inputting the original hyperspectral image into a low-rank background feature extraction network, so as to generate a low-rank background feature; inputting the low-rank background feature into a backbone feature extraction network, so as to obtain a backbone feature; inputting the backbone feature into a feature convolutional network, so as to obtain a convolution feature; inputting the convolution feature into an up-sampling network, so as to obtain an up-sampling feature; inputting the up-sampling feature into a feature convolutional sub-network, so as to obtain a reconstructed background feature map; and subtracting the reconstructed background feature map from the original hyperspectral image, so as to obtain a reconstructed error map, and using an adaptive weighted loss function to suppress an abnormal reconstruction weight. The present invention aims to solve the problem whereby low-rank characteristics of a background are not fully utilized in existing hyperspectral anomaly detection, and the problem whereby existing algorithms still face challenges in separating the background and an anomaly.

The invention discloses a laser scanning detection method for gear tooth form errors of gear motors, and belongs to the technical field of industrial optical detection. The method comprises the following steps of: firstly, comprehensively acquiring high-density three-dimensional point cloud data of a pinwheel tooth profile by combining laser scanning and precise rotating motion; secondly, preprocessing the original point cloud, and performing high-precision automatic registration on the actually measured point cloud and the standard three-dimensional CAD theoretical model by using algorithms such as an iterative nearest point; and finally, generating a visual tooth profile error map by calculating the normal deviation between the measured data and the theoretical model point by point, and quantitatively evaluating error parameters. According to the scheme, high-efficiency, high-precision and all-shape nondestructive testing of the gear tooth error is realized, physical contact damage is avoided, and the quality control level and the production efficiency of the core part of the speed reducer are remarkably improved.

The invention discloses a spaceflight propulsion chamber additive manufacturing combustion channel wall thickness uniformity control method, which comprises the following steps: S1, establishing a three-dimensional geometric model and carrying out layered slicing to generate an initial processing control parameter set; s2, constructing a thermal echo spectrum before printing the first layer, and correcting control parameters of the first layer; s3, generating a wall thickness error graph of the current layer based on the data of the previous layer, and constructing a wall thickness error potential energy field graph; s4, constructing a target scanning density function, solving a two-dimensional Monge-Amprore equation to generate a path potential function, and generating a quasi potential streamline field; s5, inputting the data of the current layer into the wall thickness prediction neural network model, and outputting a wall thickness deviation prediction value of the next layer; s6, updating the wall thickness error graph of the next layer according to the wall thickness deviation predicted value, and dynamically adjusting the control parameters; and S7, circularly executing the steps S3 to S6 until the printing task is completed. According to the method, thermal response modeling and a neural network model are fused, and dynamic uniformity control over the wall thickness of the combustion channel is achieved.

The invention discloses an environment gas leakage intelligent detection method based on big data analysis. The method comprises the following steps: collecting monitoring data and constructing a sensor network space-time diagram fused with meteorological conditions; predicting the background concentration of each node by using a sequence decomposition attention model; calculating a concentration residual error and constructing a space-time residual error map; extracting an abnormal feature vector through a physical information enhanced graph attention network; calculating an abnormal score in combination with the normal mode memory bank and generating a leakage source probability distribution diagram; when the conditions are met, calling a computational fluid dynamics model to simulate a theoretical concentration field; and determining a leakage event through spatial correlation analysis. According to the invention, the false alarm rate of gas leakage detection in an open environment is obviously reduced, and the leakage source positioning precision is improved.

The present invention discloses an error correction method and system for FPGA, which relates to the technical field of data encoding and decoding, and includes: generating Galois field elements based on the received symbol vector and the primitive polynomial, and calculating the syndrome polynomial through 8-symbol parallel processing; calculating the coefficients of the error location polynomial through the BM iterative algorithm based on the syndrome polynomial; calculating the error location by parallel substituting 10 Galois field elements through the Chien search method based on the error location polynomial; establishing the evaluation polynomial and the error pattern calculation formula, and performing polynomial division calculation through the derivative of the error location polynomial to obtain the error pattern; performing error code judgment based on the error location and the error pattern, and correcting the error symbol through the XOR operation in the Galois field to obtain the updated symbol vector. The present application adopts a parallel pipeline design structure, which enables the FPGA to automatically and real-time correct the received error module data without retransmission, effectively improving the real-time performance and reliability of error correction.

An image processing apparatus including a memory storing instructions, and a processor configured to operate based on the instructions to reconstruct a current block from encoded data of the current block, obtain a quantization error map including sample values calculated based on a quantization parameter included in the encoded data, obtain a first modified block by applying the current block and the quantization error map to a neural network, obtain a first differential block between the current block and the first modified block, obtain a second differential block by changing sample values of the first differential block based on a parameter dependent on a feature of the current block, and obtain a second modified block by combining the current block and the second differential block.

A computer-implemented method for forming a dataset configured for learning a Convolutional Neural Network (CNN) architecture including an image feature extractor. It comprises providing pairs of images, each pair comprising a reference image and a respective denoised image. For each pair of images, the method provides the pair of images to a pre-trained CNN architecture similar to the one the formed dataset will be configured for. The method computes an error map representing a difference between a first normalized feature of the denoised image and a second normalized feature of the reference image, the first and second normalized features being the output of a same layer of the pre-trained CNN architecture and adds the respective denoised image and the error map to the dataset. This constitutes an improved solution with respect to forming a dataset for learning a CNN architecture to identify areas of degradation generated by a denoiser.

A method of calibrating a robotic arm which includes perform iterative eye-in-hand and robot calibration, using a calibrated end of arm camera with known intrinsic parameters and a static target, to obtain eye-in-hand transformations and robotic parameters. The method further uses robotic parameters to estimate pose of end of arm, for eye-to-hand calibration, to obtain eye-to-hand transformations and calculates final error compensation based on the robotic parameters, eye-to-hand transformations, and eye-in-hand transformations. In one embodiment, the method calculates a robot positioning error map function, the robot positioning error map function used to adjust movement parameters for the robotic arm.

Systems, apparatus, articles of manufacture, and methods to evaluate video quality based on trained neural networks are disclosed. An example apparatus disclosed herein obtains, using a trained neural network, target features corresponding to a target video, the target features based on one or more layers of the trained neural network. The example apparatus also obtains, using the trained neural network, reference features corresponding to a reference video, the reference features based on the one or more layers of the trained neural network, the reference video associated with the target video. The example apparatus further outputs a quality metric for the target video based on the target features, the reference features, and a set of weights. In some examples, the apparatus optionally outputs an error map for the target video.

According to the thrown object detection method and device based on video anomaly detection and the storage medium, normal continuous video frames without thrown objects are used as input, then an improved auto-encoder model is trained to achieve anomaly detection of the thrown objects, multi-frame smoothing is carried out on an error graph obtained after detection, interference of a running vehicle is reduced, a brightness threshold value is set, and the thrown object detection accuracy is improved. If highlight areas exist in the smoothed error graph, the areas are contained in a minimum bounding rectangular frame, then a result is displayed and stored, it is judged that the thrown objects exist in the current time period, and meanwhile early warning is given to a system. According to the method, the detection precision of the throwing object abnormal event can be improved, a throwing object data set does not need to be marked, the problem that throwing object types cannot be exhaustive is solved, and intelligentization of road surface throwing object detection is achieved.

The invention discloses a spatial data management method and system based on digital twinning, particularly relates to the field of spatial information processing, and is used for solving the problems that a digital twinning three-dimensional model is difficult to locate defect causes in multi-source observation updating and repair is non-traceable. A target three-dimensional model is divided into surface patch units and a contribution index table is established by constructing an alignment result packet unified observation fragment set and a projection relation, a residual image is generated based on multi-view texture alignment to position a defect candidate area, a boundary projection dislocation pixel potential and a dynamic shielding coverage ratio are calculated to obtain a defect attribution result, and a defect attribution result is obtained. And performing local grid reconstruction or candidate fragment screening according to a defect attribution result, recalculating local textures to generate a repair model, updating an empirical distribution library in combination with projection re-inspection and boundary continuity inspection, and outputting a tracing record, so that integrated closed-loop treatment of defect positioning, attribution, repair and re-inspection is realized, and the consistency and credibility of the model are improved.

The invention relates to a video inter-frame decoding error image recovery method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring to-be-processed image information of a target decoding frame; wherein the target decoding frame comprises at least one decoding block; according to the to-be-processed image information, calculating area proportion data of an intra-frame prediction block in the target decoding frame to obtain frame mode attribute information of an area where a current decoding block is located; determining motion attribute information and texture attribute information of an area where the current decoding block is located according to the to-be-processed image information and the frame mode attribute information; and determining motion vector data of the current decoding block according to the frame mode attribute information, the motion attribute information and the texture attribute information, and performing image restoration on the current decoding block according to the motion vector data. By adopting the method, the motion vector calculation accuracy of the decoded error block can be improved.

The invention relates to the technical field of data management, and discloses a time-space omics experiment report generation method and device, equipment and a storage medium. The method comprises the following steps: in response to an experiment report generation instruction for a target space-time omics experiment, determining an experiment report template corresponding to the target space-time omics experiment; according to the experiment report template, reading target experiment data and an experiment image set in an experiment information storage directory corresponding to the target space-time omics experiment; scoring each experimental image in the experimental image set, and determining a target experimental image according to a scoring result; and generating an experiment report corresponding to the target space-time omics experiment according to the target experiment data, the target experiment image and the experiment report template. By automatically generating the experimental report of the space-time omics, the problems of data input errors, picture omission and the like of the experimental report caused by negligence in manual operation are avoided, and the accuracy and the integrity of the experimental report of the space-time omics are improved.

The present invention discloses a two-stage decoding method for algebraic codes, comprising: S1, performing hard decision based on a received sequence to obtain a hard decision sequence; S2, generating one or more test error pattern sequences based on the received sequence; S3, performing bitwise exclusive-OR on each error pattern sequence and the hard decision sequence to obtain a second-stage decoding input sequence; S4, determining whether second-stage algebraic code decoding is required for each second-stage decoding input sequence, and performing S5 for all input sequences requiring algebraic code decoding; S5, performing algebraic code decoding, returning a decoding status and a decoding result; and S6, selecting the optimal decoding result from the decoding output as the decoding output. The present invention can achieve performance close to the finite code length bound based on hard decision decoding when the list is small, achieving low-complexity and high-performance decoding. Furthermore, a specially introduced judgment skipping mechanism can significantly reduce the number of times the algebraic code decoder is called in the second stage, thereby significantly reducing decoding power consumption.

The embodiment of the invention provides a unified error report management method and device when multiple social media platforms send picture materials at the same time and electronic equipment, and belongs to the technical field of data processing.The method comprises the steps that an address set corresponding to a picture set P is obtained; sending the request parameter R to m social media platforms through a back-end interface API (R); constructing a binary data set Es for storing response information of the j social media platform Mj to the address URLi of the i picture; calculating a picture error rate corresponding to the social media platform Pj based on the error information Ej and the support information Sj; and calling a parameter analysis function Fj to perform parameter analysis on the error information Ej to obtain an error picture correction parameter Cg, and performing correction operation on pictures in an error picture set Pcj corresponding to the social media platform Pj based on the correction parameter Cg and a picture correction function Fg. According to the invention, the error correction efficiency of multi-platform picture sending can be improved.

The invention relates to the technical field of data analysis, and provides an error analysis method and system, electronic equipment and a storage medium, and the method comprises the steps: carrying out the preprocessing of to-be-analyzed data, and obtaining an attribute vector; based on the analysis path, identifying errors in the attribute vector to obtain an error identification result; carrying out error classification on the error identification result, analyzing true errors in an error classification result, and determining error causes; and displaying the error reasons to obtain an error map, and displaying the analysis process of the true error in the form of a tree diagram to obtain an analysis map. The method supports a user to customize an analysis path, can improve the flexibility of error analysis and meet the requirements of the user, can improve the efficiency and accuracy of error analysis by classifying error recognition results and focusing on analysis of true errors, and can improve the accuracy of error analysis by displaying an error map and an analysis map. A user can more intuitively and quickly position the error reason, so that the analysis efficiency is improved.

The invention provides an image forgery detection method and system based on asymmetric anchoring of a multi-modal large model, and belongs to the technical field of image detection processing. Truth value anchor point features and to-be-optimized intermediate features of an input image are extracted; an asymmetric strategy is executed based on the authenticity category of the image, and anchoring alignment loss is introduced only for the real image to anchor real feature distribution; extracting global semantic features, and mapping the global semantic features into the authenticity category probability of the image through a classification projection head; and generating a space alignment error graph by using the deviation between the intermediate feature to be optimized and the true value anchor point feature, injecting the space alignment error graph into a positioning decoder, and generating a forged region mask by combining the output feature of the positioning encoder. According to the method, a true value anchor point is redefined, and an asymmetric anchoring mechanism is utilized to force real image features to return to real world priori, so that the problem of characterization drift is solved; and generating a spatial error graph by using the feature alignment deviation, thereby realizing stable learning of forged clues and realizing positioning of an accurate tampering region.

To quickly specify the cause of a failure related to a controller and a system composed of a plurality of controllers on the basis of a log.SOLUTION: The map generation unit 103 generates, for each of the plurality of logs acquired by the log acquisition unit 102, a log map including the plurality of acquired logs, in which the type of the log and the code of the log are associated with the coordinates of the bitmap data to form a bit of the bitmap data, and the importance generated by the importance generation unit 101 is used as color information of the bit. The estimation unit 105 estimates an abnormality in control by the controller 121 based on the result of pattern matching between the log map generated by the map generation unit 103 and the error map stored in the error storage unit 104.SELECTED DRAWING: Figure 1

The embodiment of the present invention provides a method for detecting image mosaic misalignment points based on satellite images and VGG model fine-tuning, the method comprising: preprocessing mosaic images of satellite images, performing misalignment judgment, and determining misalignment areas; marking the misalignment areas of the mosaic images, and dividing the marked mosaic images into a training set and a verification set; inputting the training set into a VGG model for training, calculating a loss function through a predicted image obtained through VGG model training and the training set, and training until the VGG model is stable; inputting the verification set into the trained VGG model, comparing the output image of the VGG model with the verification set, and evaluating the correctness of the model output result; when the correctness is less than a preset accuracy value, re-marking the erroneous image, iteratively training the VGG model through the re-marked erroneous image, and outputting the model weight of the VGG model; integrating the model weight into a preset detection script, and detecting the mosaic misalignment points of the original satellite image in real time.

The invention belongs to the technical field of unmanned aerial vehicle take-off and landing guidance, and discloses an unmanned aerial vehicle take-off and landing guidance radar data correction method based on an error map. Comprising the following steps: S1, acquiring a reference data stream of the airborne RTK / INS integrated navigation system; an unmanned aerial vehicle take-off and landing guide radar data stream is acquired; obtaining an error set of the reference data stream and the radar data stream; s2, establishing an error map according to the error set; generating a three-dimensional error distribution diagram of the unmanned aerial vehicle take-off and landing guide radar in the airport airspace according to the error map And S3, integrating the three-dimensional error distribution map as a lookup table into an adaptive filter based on an aircraft kinematics model, performing real-time table lookup according to the current estimated position of the aircraft to obtain an expected system deviation, and performing compensation in an updating step.

A radar signal processing device includes: a Fourier transform unit that Fourier transforms an image signal indicating a synthetic aperture radar image in an azimuth direction; and a map division unit that divides in a Doppler frequency domain a range Doppler frequency map indicated by a signal after the Fourier transform, and outputs signals that are a plurality of the divided range Doppler frequency maps; an inverse Fourier transform unit that inverse Fourier transforms signals indicating the divided maps output, and outputs divided playback signals that are signals after the inverse Fourier transform; a gain calculation unit that calculates a false image suppression gain for suppressing a false image of a target appearing in the synthetic aperture radar image using the divided playback signals output and the image signal; and a false image suppression unit that multiplies the image signal with the false image suppression gain calculated.

The present invention relates to a computer-implemented method 1000 for identifying patterns (23) that correlate with correct image classifications (52) and incorrect image classifications (52) using a scene graph (12). [Solution] For a set of images (11), this method obtains a scene graph (12) from the images, classifies the images using a pre-trained classifier (5), groups the scene graphs (12') according to whether the classification is correct, extracts representative subgraphs within each group, thereby revealing patterns associated with the classification (52).

The present invention is for, in applying a target object exploration according to a RPA scenario based on a previous UI screen to the current UI screen, if a specific target object cannot be searched or an error occurs due to a change on a target app / web at the current time point, automatically extracting information on the most suitable corresponding object on the current UI screen, thereby automatically recovering from the target object exploration execution failure of the previous scenario. A method of the present invention may comprise: an error recognition step for detecting the occurrence of an error resulting from the application, to the current UI screen, of a target object exploration based on object exploration information of a RPA scenario created on the basis of a previous UI screen; an image information LLM inquiry step for generating a first prompt including a first image including at least a portion of the previous UI screen including a target object, and a second image including at least a portion of the current UI screen, and instructing a generative AI to return information of an object corresponding to the target object on the current UI screen; and an object exploration step for performing target object exploration on the current UI screen.

The present invention provides a method and device for adaptive sampling of magnetic resonance signals based on deep learning, comprising: determining a sampling space, a sampling value threshold, and the number of sampling points for a single excitation; selecting an initial region to be sampled based on the number of sampling points; determining a 3D epi-intensity spectroscopy (EPI) sequence corresponding to a single excitation based on phase encoding of the sampling points within the region to be sampled; performing a single radiofrequency excitation based on the 3D EPI sequence to obtain a corresponding gradient echo signal; reconstructing an optimized dMRI image based on the gradient echo signal; inputting the optimized dMRI image into a trained full-sampling dMRI image prediction model to predict a full-sampling dMRI image and obtain a reconstructed image error map; obtaining a sampling value for each sampling point in the sampling space based on the reconstructed image error map; and determining a region to be sampled for the next sampling based on the sampling value of each sampling point and the sampling value threshold. This application further shortens scanning time and improves the accuracy of the generated dMRI images.

A method and apparatus for adaptive machining of freeform surfaces, the method comprising: acquiring a CAD model of a freeform surface workpiece; generating a machining path and measurement information for a blank workpiece based on the CAD model; sending measurement control commands to a coordinate measuring machine (CMM) based on the measurement information; measuring the blank workpiece clamped on a CNC machine tool based on the machining path; and obtaining coordinate data P for each target measurement point. me According to the CAD model and P me The machining accuracy of the blank workpiece is judged. When the machining accuracy is less than a preset threshold, an ideal semi-finished CAD model of the freeform surface workpiece is obtained. A mirror M is constructed between the CAD model and the ideal semi-finished CAD model. m Based on P me and M m Get each P me Corresponding reflection point coordinate data P com Based on all P com Obtaining an error map and using it as a finishing path for finishing workpieces can ensure machining accuracy, improve machining efficiency, and reduce machining costs during the machining process.

The application relates to the technical field of decoding, and provides a decoding method, a decoding device, electronic equipment and a storage medium, wherein the method comprises the following steps: determining an error pattern of each check equation in a check equation set based on a hard decision result, determining the modulus of a bit position in a received sequence, determining a bit flip sequence corresponding to the received sequence based on the modulus of the bit position and the error pattern, determining an original code word estimation based on the bit flip sequence and the hard decision result, and determining whether the original code word estimation satisfies all check equations in the check equation set. In the method, the modulus of the bit position provides reliability information of each bit position, and the error pattern is used for reflecting the inconsistency between the received sequence and a check matrix, so that the decoding algorithm can correct errors in a targeted manner; in the decoding process, decoding based on the two can more accurately reflect the degree of influence of each information on noise, thereby reducing the bit error rate of the final decoding result and guaranteeing the accuracy and integrity of communication data.

A method for decoding an image according to one embodiment may comprise the steps of: determining a class of the current block on the basis of the amount of change in sample value for each sample in the current block; determining, on the basis of the class of the current block, a multi-channel kernel to be applied to the current block; determining a feature vector for the current sample by applying each of a plurality of filter kernels included in the multi-channel kernel to the current sample in the current block; acquiring an expected error mapped by the feature vector from a predefined data set; and filtering the current sample by using the expected error.

The invention provides an intelligent customer service dialogue generation and optimization method and system based on a large model. According to the method, data such as repeated questions, manual rewriting, cursor tracks, text deletion and voice interruption of a user are collected, and timestamps are marked. And calculating an operation density value and an error frequency value to generate a corrected heat map. And defining a strength node and a triggering condition edge according to the dynamic error graph. The dynamic error graph is updated when data are newly added, and an adversarial sample is generated based on the characteristics of the dynamic error graph. And the large model response module is finely tuned through reinforcement learning negative feedback, optimization content is output, the updated atlas is fed back, and closed-loop iterative optimization is realized. According to the method, a dynamic error graph is constructed by utilizing user interaction data to generate an adversarial sample, a large model is driven to perform closed-loop optimization on customer service dialogue generation through reinforcement learning, and errors are accurately positioned and continuously improved.

The invention relates to the field of wear monitoring, in particular to a cross guide rail wear state monitoring method based on machine vision, which comprises the following steps: acquiring an initial image and a standard image of a cross guide rail, subtracting the standard image from the initial image to obtain an error image, performing binarization processing on the error image by using an Otsu threshold method to obtain a binary image, and calculating the wear state of the cross guide rail; taking an area formed by white pixel points in the binary image as an abnormal area; calculating a texture complex factor and a shape factor of the abnormal area, and normalizing the product of the texture complex factor and the shape factor to obtain a wear characteristic value; and extracting an edge line of the abnormal area, calculating a mean value of gradient amplitudes of pixel points of the edge line, and carrying out weighted summation on the wear characteristic value and the mean value of the gradient amplitudes of the pixel points of the edge line to obtain a wear index for judging whether the surface of the cross guide rail is worn or not. According to the invention, the accuracy of cross guide rail state monitoring is improved.

The invention discloses a no-reference image quality evaluation method based on a prediction error graph, which is suitable for the field of image processing, realizes the evaluation of image quality through a deep learning model, and comprises the following steps: constructing a prediction error graph pre-training model based on a Transform encoder and decoder structure; inputting the distorted image into a prediction error graph pre-training model to generate a corresponding prediction error graph: constructing a stepped feature extraction network based on decomposed large kernel convolution, performing layer-by-layer feature extraction on the distorted image, obtaining multi-scale features, and aggregating the multi-scale features to form global features; carrying out element-by-element fusion on the extracted image features and the features of the prediction error graph; and finally, the fused features are mapped into image quality scores. According to the method, the image distortion region and the degradation mode thereof are described by introducing the prediction error graph, and the prediction error graph is combined with the multi-scale features, so that the model can fully utilize the image degradation information, and the accuracy and the stability of no-reference image quality evaluation are improved.