131 results about "Global matching" patented technology

The invention relates to the technical field of positioning, in particular to a device based on vision and 2D laser fusion positioning, and the device comprises a horizontal distance measurement module which is horizontally installed on a robot body and is used for collecting contour distance measurement data of a motion plane; the vertical vision module is installed on the robot body in a top view mode and used for collecting visual feature data of a vertical space; the calculation processing module is configured to construct a laser contour map and a visual feature map which are aligned in a coordinate system based on the contour distance measurement data and the visual feature data, determine an initial pose of the robot according to a given position or an optimal position of self-global matching of the laser contour map and the visual feature map, generate a predicted pose in combination with a motion measurement value, and send the predicted pose to the robot. And iteratively executing dual-mode observation filtering updating of laser contour matching optimization and visual reprojection matching optimization by taking the predicted pose as a center. According to the invention, through deep integration of a heterogeneous sensor architecture and a computing platform, an industrial scene-oriented all-weather positioning solution is constructed.

The invention discloses a crop disease multi-modal diagnosis and classification method and system, and relates to the technical field of data processing, and the method comprises the steps: collecting disease multi-modal data, and carrying out the preprocessing; constructing a disease multi-modal diagnosis model, respectively inputting images and text sequences in the multi-modal data into a visual feature extraction subnet and a text feature coding subnet of the disease multi-modal diagnosis model, and extracting global visual features, sequence state features and global text features; calculating a global matching score based on the global visual features and the global text features, executing fine-grained local interaction on the global visual features and the sequence state features, introducing a category channel attention mechanism to correct the features obtained by interaction, and weighting to generate multi-modal fusion features; and constructing a loss function based on the global visual features, the global text features and the multi-modal fusion features, training a disease multi-modal diagnosis model, inputting to-be-classified data into the disease multi-modal diagnosis model, and outputting a classification result.

The invention discloses a method and a system for detecting residues in a closed steel drum based on visual inspection, and belongs to the technical field of industrial nondestructive inspection and computer vision crossing, and the method comprises the following steps: collecting a multi-angle image through dynamic light field regulation and control, and generating a high dynamic range image; constructing a dense three-dimensional point cloud by applying a semi-global matching algorithm, and fusing the infrared heat absorption coefficient and the visible light texture roughness to generate a three-dimensional model with material attributes; based on structural features, filtering out potential targets in a non-residual region, and combining point cloud height variance and HSV hue gradient analysis to generate matter state discrimination parameters; training a convolutional neural network by using the voxelization model and the circumferential weld feature template, and outputting an anti-interference residue segmentation mask; a two-dimensional projection drawing is obtained through cylindrical coordinate expansion, and the residue weight is calculated by combining a density mapping library. According to the method, a dynamic light field regulation and control and multispectral fusion technology is adopted, and a three-dimensional point cloud reconstruction and deep learning algorithm is combined, so that accurate detection and quantitative analysis of the residues in the closed steel drum can be realized.

The invention discloses a binocular camera global ultra-sensitive depth direction vibration measurement method, and relates to the technical field of structural vibration measurement. The method comprises the following steps: completing internal reference and external reference calibration of a binocular camera based on a Zhang's calibration algorithm, and ensuring the three-dimensional reconstruction precision; in image processing, effective pixels with high dynamic response are screened by calculating pixel intensity time-history variance, a pixel intensity space-time matrix is constructed, and singular value decomposition is adopted to extract front k-order modals to reconstruct the pixel intensity matrix so as to reveal real signals originally submerged in imaging noise and rounding errors. And multi-direction cost aggregation is carried out by using a semi-global matching algorithm, the parallax calculation robustness in a low-contrast scene is improved, and a high-precision depth map is generated. And depth information is converted into three-dimensional space information in combination with camera parameters, so that non-contact measurement of the vibration time travel curve in the full-field depth direction is realized.

The invention discloses a fuzzy video real-time jitter removing method and system, and the method comprises the steps: distributing feature points through multi-scale FAST detection, lightweight MLP detection and dynamic weight, combining a MobileNet-FPN network and the lightweight MLP dynamic weight to extract the robust features of the feature points, modeling the global matching relation of the feature points of two adjacent frames through Transform, and carrying out the real-time jitter removing of a fuzzy video. And a Bayesian adaptive RANSAC algorithm is adopted to realize high-efficiency mismatching pair elimination, and finally an affine transformation matrix is solved and reverse mapping is performed to compensate jitter. According to the method, while the real-time performance of the mobile terminal is ensured, the image stabilization precision in a violent shaking scene is remarkably improved.

The invention discloses a real-time multi-target tracking method based on occlusion information. The real-time multi-target tracking method specifically comprises the following steps: step 1, initializing a tracker; step 2, obtaining a current frame target and appearance characteristics of the current frame target, and sending the obtained target and corresponding characteristics into a tracker; 3, constructing an incidence matrix, and carrying out two-stage global matching to obtain a matching result M; step 4, correcting the influenced target based on the division of the track category in the step 1 and the division of the target category in the step 3; then the track information is updated; 5, increasing the shielding duration of the unmatched tracks, deleting the tracks with overlong shielding duration, converting the types of the tracks, and initializing the unmatched targets; and step 6, after track category conversion is completed, outputting a current frame tracking result, and returning to the step 2 for cyclic tracking until tracking is finished. According to the method, the tracking effect on the low-confidence target in the low-camera-angle scene is effectively improved, and the accuracy of the multi-target tracking task is improved.

The invention discloses a method for realizing intelligent sag measurement of a power transmission line by using a binocular depth camera, which comprises the following steps of: firstly, fixing the binocular camera on a rotary holder at the central position of two towers to complete camera calibration and epipolar correction; then, controlling the holder to rotate according to a fixed stepping angle, collecting a binocular image sequence covering the whole section of lead, and preprocessing the binocular image sequence; performing feature matching and image splicing by using an SIFT algorithm based on the left eye image sequence to generate a panorama and establish a panorama-local mapping relationship; thirdly, detecting an insulator and a spacer target in the panorama based on a Mask-RCNN model, accurately positioning key point coordinates of the insulator and the spacer target, and mapping the key point coordinates to an original local image; in the corresponding left and right local images, a semi-global matching (SGM) algorithm is adopted to calculate parallax and generate depth information, and then three-dimensional coordinates of a target point are obtained through conversion; and finally, performing catenary fitting according to the ordered three-dimensional point set, introducing a temperature and windage yaw dynamic correction model, and accurately calculating a sag value. According to the invention, non-contact, automatic and high-precision sag measurement is realized, the method is suitable for routing inspection of complex terrains and large-span power transmission lines, and the operation safety and efficiency are remarkably improved. The invention also provides a system, equipment and a medium for realizing the method.

The invention provides a silk-screen defect detection method and device, electronic equipment and a storage medium, and belongs to the technical field of intelligent manufacturing, and the method comprises the steps: carrying out the global matching of a to-be-detected image of a to-be-detected product through a global template; determining an initial matching position of the local template in the to-be-detected image based on the global matching result and the relative position relationship, determining a local search area according to the initial matching position, and performing accurate matching in the local search area to obtain a to-be-detected area image accurately aligned with the local template; and comparing the to-be-detected area image with the local template to generate a silk-screen defect detection result. According to the invention, global matching is carried out by using the global template, and then each local template is accurately aligned in the local area of the to-be-detected image based on the global matching result and the relative position relation of the local template relative to the global template, so that false alarms caused by inaccurate alignment are reduced, and the efficiency and accuracy of silk-screen defect detection are improved.

The invention provides a guide registration method, system and device for RGB and infrared image fusion and a storage medium, and belongs to the technical field of image enhancement based on computer vision. Firstly, input of a visible light image and an infrared image is obtained, main target objects in the images are recognized and detected, and a preliminary target detection frame is generated; then, taking the target detection frame as spatial prior information, and generating an RGB image segmentation result and an infrared image segmentation result; performing feature matching on two segmented image results through a newly-designed cross-modal feature matching engine strategy, firstly extracting coarse-grained and fine-grained features through two parallel encoders, then obtaining a homography matrix through a global matcher, distortion refining and matching sampling, and finally obtaining a cross-modal feature matching algorithm; and then homography transformation is carried out on the original infrared image to obtain a final registration image. The method is superior to an existing method in the aspects of registration accuracy and stability, and a robust solution is provided for registration of infrared and visible light images.

The invention relates to the technical field of geophysical exploration, and discloses a three-dimensional fine electromagnetic detection method for ore control fracture of a deep gold ore. The method comprises the following steps: arranging a measuring point array to emit a coding artificial electromagnetic pulse signal, and synchronously receiving a plurality of induced magnetic field responses to form an original data volume; estimating a noise baseline based on statistical distribution and adaptively calibrating data; constructing a spatial correlation network by using the time sequence correlation of adjacent measuring points, and identifying a high-connectivity cluster as a suspected fracture region; extracting a regional time sequence to perform multi-level clustering, and screening a typical fracture response mode; based on the mode set global matching enhancement response, generating an enhancement data volume; multi-scale three-dimensional tensor decomposition is adopted to extract a tensor kernel component, and an electrical structure is reconstructed according to the spatial compactness and energy attenuation characteristics of the tensor kernel component; and finally, iteratively fusing with a geological constraint model to output a three-dimensional ore control fracture fine detection result. According to the method, intelligent identification and high-precision imaging of the deep fracture are realized.

Examples relate to a concept for an entry-exit matching system, and in particular to an evaluation device, a method and a computer program for person re-identification for entry-exit matching in a transportation system. The evaluation device comprises processing circuitry configured to obtain a plurality of re-identification codes. Each re-identification code represents a person being recorded by at least one camera when entering or exiting at least a section of the transportation system. The processing circuitry is configured to match the plurality of re-identification codes using a global matching scheme to obtain a plurality of matched pairs of re-identification codes, such that each matched pair of re-identification codes comprises a re-identification code of a person entering and a re-identification code of a person exiting. The global matching scheme is based on reducing an overall distance between the re-identification codes of the matched pairs of re-identification codes over the plurality of matched pairs of re-identification codes. The processing circuitry is configured to determine points of entry and exit for the plurality of matched pairs of re-identification codes.

The invention relates to the technical field of three-dimensional modeling, in particular to an augmented reality-oriented three-dimensional scene reconstruction method and system, and is used for solving the technical problem that an iterative nearest point algorithm cannot meet high real-time performance and high-precision reconstruction requirements required by augmented reality at the same time. The method comprises the following steps: acquiring a pose transformation relationship between multiple frames of point cloud data and adjacent frames of point cloud data; extracting a plurality of feature points in each frame of point cloud data, and determining a feature descriptor of each feature point; according to the feature descriptors of the feature points, calculating a matching weight between feature point pairs in two adjacent frames of point cloud data, and calculating a local matching error based on the matching weight and a pose transformation relationship; performing global matching optimization based on the local matching error of the plurality of feature point pairs and the distribution density of each feature point in the respective point cloud frame, and determining an optimal matching point pair set which minimizes the global matching error; and according to the optimal matching point pair set, fusing the multi-frame point cloud data, and reconstructing a three-dimensional scene model.

The invention discloses a future computing-oriented super-node cluster liquid cooling efficient computing platform and a resource management method thereof. A CPU, a GPU and a reserved photon / quantum accelerator slot in a super node are virtualized into a computing resource pool through a unified heterogeneous resource abstraction layer, and a CXL large-capacity memory and an NVMe-oF storage are virtualized into a cross-node memory resource pool and a storage resource pool respectively; the physical state of the super node cluster, the temperature-flow distribution of the liquid cooling heat dissipation system and the operation parameters of the heat recovery system are mapped in real time in the digital twin system; the intelligent scheduler takes task completion time, cluster total power consumption and waste heat recovery efficiency as a joint optimization target, and decomposes the task based on a virtual model; and in the task execution process, the energy efficiency cooperative controller adjusts the operation frequency of the allocated resources and the rotating speed of the liquid cooling pump in real time according to the node temperature change predicted by the digital twin model, so that the task performance, the energy consumption and the heat dissipation capability are globally matched.

Apparatuses, systems, and techniques to determine optical flow. In at least one embodiment, a set of disparity values is used to determine optical flow between input and reference images. For each of a plurality of image regions of the input image, the set of disparity values may include disparity values for a plurality of directions intersecting the image region.

The invention provides a robot repositioning method in a dynamic environment based on laser visual complementation. The robot repositioning method comprises two stages of offline mapping and online repositioning. In the off-line stage, two-dimensional and three-dimensional elimination of a dynamic target is realized through a lightweight detection network and geometric projection, and the purity of a prior map is improved; visual and laser radar static features are respectively extracted, a complementary double descriptor database is constructed, and laser radar descriptors fuse geometry and reflection intensity information to enhance discrimination. In the online stage, a coarse-to-fine strategy is adopted, rapid and robust global matching is achieved by combining GPS coarse positioning and radius search-based database retrieval, and scene recognition and accurate pose recovery are completed by using vision and radar dual channels. According to the method, a high-purity priori library is constructed through dynamic target collaborative elimination and multi-modal feature fusion, and autonomous high-precision global repositioning without manual initial pose input is realized in combination with GPS coarse positioning and a self-adaptive channel switching mechanism.

The application discloses a kind of crop disease multi-modal diagnosis and classification method and system, it is related to data processing technical field, the method of the present application includes collecting disease multi-modal data and pre-processing;Disease multi-modal diagnosis model is constructed, the image and text sequence in multi-modal data are respectively input into the visual feature extraction subnet and text feature coding subnet of disease multi-modal diagnosis model, extract global visual feature, sequence state feature and global text feature;Based on global visual feature and global text feature, calculate global matching score, perform fine-grained local interaction on global visual feature and sequence state feature, and introduce class channel attention mechanism to correct the features obtained by interaction, and generate multi-modal fusion features by weighting;Based on global visual feature, global text feature and multi-modal fusion feature, construct loss function and train disease multi-modal diagnosis model, input the data to be classified into disease multi-modal diagnosis model and output classification results.

This invention discloses a long text matching method combining noise filtering and a divide-and-conquer strategy. The method includes: constructing a long text matching model, which comprises a keyword extraction layer, an association extraction layer, and a filtering layer. The keyword extraction layer extracts keywords from the text; the filtering layer filters noise from the text based on sentence similarity to obtain a denoised text sequence; and the association extraction layer further removes keywords from the denoised text sequence to obtain the remaining associated text. The long text matching model is trained with the optimization objective of minimizing a set overall loss function, which reflects the global matching distribution and combines the keyword and association matching distributions. For the target text, real-time text matching is performed using the trained long text matching model. This invention improves the generalization ability and accuracy of text matching.

This invention discloses a video tag dynamic overlay anti-shake method and system based on AR, comprising: acquiring real-time video stream and device pose parameters at corresponding timestamps; constructing a dynamic perspective projection model based on the device pose parameters and calculating the initial pixel coordinates of the target AR tag in the current video frame; planning a local feature search window based on the initial pixel coordinates; calculating the average pixel displacement vector of image feature points in the local feature search window of the current and adjacent historical video frames; performing reverse displacement compensation on the average pixel displacement vector to obtain the target rendering coordinates, and rendering the target AR tag to the current video frame based on the target rendering coordinates. Compared with the prior art, this invention can overcome the shortcomings of large drift errors caused by simply relying on mechanical pose, high computational power consumption caused by global matching, and high-frequency jitter of tags caused by slight device vibrations, thereby improving tag mapping accuracy, reducing system computational power consumption, and improving the visual stability of video overlay.

This invention proposes a method and system for detecting missing weld points on vehicle bodies based on 3D vision and global matching, belonging to the field of image recognition technology. The method includes: determining the transformation relationship between the camera coordinate system and the vehicle body coordinate system through calibration; at each observation position, a 3D camera acquires an image I containing a frame of 2D texture and the corresponding 3D data; inputting the acquired image I into a pre-trained deep learning object detection model, outputting the bounding box information (u, v, w, h) for each weld point; for each detected weld point, calculating the set S of the weld point's 3D coordinates in the vehicle body coordinate system. det For the theoretical solder joint library S cad Each theoretical solder joint P in cad,i In set S det The invention employs a "detect first, match later" strategy, which does not rely on the precise projection of the theoretical position, thus reducing the requirements for the positioning accuracy of the production line tooling fixtures and conveying system.

The present invention discloses a method for detecting contraband in X-ray security images based on generative sparsity. The method first obtains an X-ray security image dataset containing a target bounding box and category annotations, and successively uses ResNet and a feature pyramid structure to extract multi-level features in the security image. Secondly, a diffusion model is used to generate a noisy plan frame, and the plan frame is used to generate features of the region of interest and a feature representation with context awareness; a refined feature map is generated through a dynamic filtering and fusion module to obtain a target prediction frame and a corresponding category score. Then, in the label assignment stage, the Hungarian Top‑K matching strategy is used to find K equally important optimal positive samples for each true value through global matching, and a loss function is constructed to perform iterative training. The training is verified on a validation set, and a contraband detection effect map is output. The present invention performs well both in terms of X-ray contraband detection accuracy and timeliness.

The present invention belongs to the technical field of object tracking in computer vision, and discloses a drone object tracking method, system, medium, device and terminal. The backbone network is pre-trained using the ImageNet dataset; a SiamTCP drone tracking model is created, and the SiamTCP drone tracking model is trained on multiple video datasets. The drone object tracking method based on target-aware cascaded pixel matching provided by the present invention determines the parameters of the neural network model by training on multiple datasets. The trained model can be applied to the object tracking task in the drone scenario. Specifically, the present invention proposes pixel-by-pixel cross-correlation and performs a cascaded operation, replacing the global-to-global matching with a part-to-part matching operation, enhancing the result of feature fusion. At the same time, the target-aware module proposed by the present invention can eliminate unnecessary background information and can cooperate well with the cascaded pixel matching process.

The application relates to a multi-sensor-based repositioning method, device, equipment and storage medium. The method comprises the following steps: acquiring environment images collected by respective sensors; performing global matching on the respective environment images and images in a prior environment image library to obtain respective robot poses before updating corresponding to the respective sensors; performing local matching on the respective environment images and respective surrounding environment images corresponding to the respective robot poses before updating in the prior environment image library to obtain score values of respective robot poses after updating; and determining a repositioning result of a current robot pose according to the score values of the respective robot poses after updating. By performing global matching and local matching on the environment images collected by different types of sensors and the images in the prior environment image library, the repositioning result of the current robot pose can be ensured to have accuracy and reliability.

The application relates to an ionosphere virtual height estimation method applied to a short-wave active-passive integrated radar, which comprises the following steps: 1: an integrated detection system receives active radar detection echoes and passive radar detection echoes, and a main-passive detection channel is respectively subjected to distance, speed and azimuth sorting to obtain an active radar point track measurement set and a passive radar point track measurement set; 2: a group of points is respectively taken out from the two measurement sets, and the ionosphere virtual height is calculated by assuming the points as matched points; 3: the active radar measurement and the passive radar measurement are converted to the northeast sky coordinate system according to the calculated ionosphere virtual height; 4: a threshold method is used for global matching on the converted measurement sets to obtain a matching result, and a cost function value is calculated according to the matching result; 5: the two measurement sets are respectively traversed, and the steps 2, 3 and 4 are repeated, and the ionosphere virtual height corresponding to the minimum cost function is taken as the ionosphere virtual height in a detection environment. The application is used in the field of high-frequency over-the-horizon radar signal processing.

Establishing an AI empowered global loan matching of borrowers and lenders, by creating a loanchain of “atomic loans”. Large long duration loans are constructed from a succession of small, short duration loans, thereby mitigating lender's risk, and pulling into the loan dynamics hundreds of billions of dollars that currently earn no interest to their owners.

The present invention relates to the technical field of remote sensing image semantic segmentation, and in particular to a remote sensing image semantic segmentation method and system based on prototype perception and edge enhancement. The method comprises: inputting K supporting images and a query image into a trained remote sensing image semantic segmentation model to obtain a semantic segmentation result of the query image; wherein the model comprises: performing feature extraction on the K supporting images to obtain supporting features; extracting a plurality of prototypes from the supporting features; performing feature extraction on the query image to obtain query features; performing pixel-level feature matching on the query features based on the prototypes to obtain global matching features; performing edge information extraction on the query image and performing feature enhancement on the query features based on the extracted edge information to obtain edge enhancement features; performing feature fusion on the query features, the global matching features and the edge enhancement features to obtain fused features, and predicting the fused features to obtain an image segmentation result.

The invention provides a semi-global matching binocular depth estimation method based on polarization information guidance, and the method comprises the steps: obtaining a focal plane polarization optical image of a target under a binocular visual angle, calculating the polarization information of the focal plane polarization optical image, and obtaining the polarization degree information and the polarization angle information; constructing a polarization matching cost function by adopting a window matching method based on the polarization angle information; the method comprises the following steps: constructing a gray scale matching cost function by adopting an SAD-Census hybrid calculation method based on a focal plane polarization optical image; taking the polarization degree information as a fusion weight, and carrying out weighted fusion processing on the polarization matching cost function and the gray matching cost function to obtain a fusion cost function; performing semi-global matching SGM based on polarization clue guidance by adopting a fusion cost function to obtain final parallax information of the target; and performing three-dimensional reconstruction processing on the target through the final parallax information of the target to obtain target point cloud data. On the basis, the matching precision and the robustness are remarkably improved.

This invention proposes a hinged vehicle trajectory planning method based on multi-camera surround view information. It estimates the relative pose of the cameras based on the surround view images, combines the independent semantic cognition of each camera to obtain a bird's-eye view semantic cognition map, and establishes a time-invariant global matching relationship with the semantic map. Based on the matching features, it completes high-precision estimation of the vehicle pose. By improving the accuracy of vehicle pose estimation and the long-term robustness of the matching features, it enhances environmental perception and localization accuracy, providing high-precision data to be fused for subsequent trajectory planning. This invention designs a vehicle trajectory planning method based on the bird's-eye view semantic cognition map and vehicle pose estimation. It fuses the constructed traffic strategy set with quasi-static representation and vehicle pose to complete trajectory planning, improving the real-time performance and completeness of trajectory planning. Tracking is performed based on the predicted trajectory, and the traffic strategy set is updated based on the tracking accuracy, improving the feasibility of trajectory planning.

The application discloses a kind of multilingual contract intelligent comparison method, system, equipment and storage medium, and its technical scheme points are in that first language version contract and second language version contract are preprocessed to obtain first text content and second text content;First text content and second text content are corrected to obtain first corrected text and second corrected text;All clauses in first corrected text and second corrected text are traversed, the overall semantic similarity and multilevel semantic similarity between the clauses are calculated, and the global matching degree between them is calculated;According to overall semantic similarity, multilevel semantic similarity and global matching degree, the similarity score between the clauses is calculated, and whether the clause is a problem clause is determined according to the similarity score;Identify the problem category of problem clause, and generate corresponding correction suggestion according to the problem category.The application realizes automatic comparison of multilingual contract, and ensures the consistency and accuracy of multilingual contract.

The present application relates to a kind of power transmission engineering smoke hidden danger identification method, device and storage medium.In the present application, the output feature of the motion recognition of the motion recognition network of the DBG video action recognition network is cascaded with the output optical flow feature of the global matching optical flow network of GMFlow after operation, and then input into the first Transform for deep feature fusion;In the first Transform, the feature map after target feature extraction is converted into input-output sequence, and then the multiple self-attention layers are optimized to calculate the multiple attention outputs linearly connected to the desired dimension;Obtain the hidden danger data containing smoke hidden danger in power transmission engineering, train the recognition network architecture by hidden danger data to obtain parameters, and finally obtain the recognition model for power transmission engineering smoke hidden danger.The present application detects smoke by combining spatial features and time sequence features, greatly reduces the frequency of smoke false alarm and improves the accuracy of prediction.