144 results about "Location map" patented technology

The invention discloses a dynamic Gaussian digital human image rendering method and device, equipment and a storage medium, and relates to the technical field of artificial intelligence. The method comprises the following steps: determining a target multi-view image frame from a multi-view RGB video image sequence according to a preset attitude, and constructing a deformable parameterized model according to the target multi-view image frame; constructing an attitude space driving attitude corresponding to the multi-view RGB video image sequence based on a three-dimensional attitude estimation technology, and generating a two-dimensional position map according to the attitude space driving attitude and the deformable parameterized model; performing three-dimensional Gaussian binding on the deformable parameterized model to obtain a local attribute of the three-dimensional Gaussian; training is carried out according to the two-dimensional position map, and a target StyleUNet neural network is obtained; and predicting the new attitude through the target StyleUNet neural network to obtain a dynamic Gaussian digital human image. In this way, the high-fidelity drivable high-frequency detail digital human image can be automatically rendered.

Aspects of the subject disclosure may include, for example, accessing device usage data that can be generated when end user devices connect with the network; and analyzing the device usage data according to identification information of the deployable cell site, which enables determining or estimating a location of the deployable cell site. In one embodiment, this analysis can be based on device usage reports and network performance data. A performance report and / or location map can be generated for an area associated with the deployable cell site. Other embodiments are disclosed.

The invention discloses a stacker walking type RFID rapid checking method, device and equipment, and relates to the technical field of intelligent warehousing automatic checking, and the method comprises the following steps: obtaining a position sequence of a stacker on a walking path, and synchronously analyzing the recognition state of an RFID tag to obtain a tag reading position map; based on the tag reading position map, identifying a critical point of identification state change of the RFID tag in the position sequence; segmenting the walking path based on the critical point to obtain a tag identification section, and performing center offset evaluation on the tag identification section to obtain an optimal identification position point of the RFID tag; based on the optimal identification position point of the RFID tag, the polling trigger sequence is constructed, the RFID reader-writer is controlled to execute tag polling according to the polling trigger sequence, a final inventory list is obtained, and the problem that the tag frequently quits and joins identification polling due to dramatic fluctuation of tag reflection intensity in the walking process of the stacker is solved.

A method includes, detecting a pointing motion made by an occupant of a vehicle and causing a photograph to be taken by a camera associated with the vehicle. The method also includes processing one or more of the photograph, a time the photograph was taken, location information of the vehicle at the time the photograph was taken, a photographing direction extending from the camera at the time the photograph was taken, or map data to obtain a candidate object of interest in the photograph and description information associated with the candidate object of interest. The method further includes causing a graphical user interface to be output by a display. The graphical user interface includes a geographical map, a location icon indicating a location on the geographical map at which the photograph was taken, a graphical object representing the photographing direction, and description information associated with the candidate object of interest.

One embodiment of the present invention sets forth a technique for generating a geometry for a shape. The technique includes inputting, into a machine learning model, (i) a noise sample and (ii) one or more conditioning inputs. The technique also includes generating, via execution of the machine learning model based on the noise sample and the one or more conditioning inputs, a two-dimensional (2D) position map associated with the shape. The technique further includes generating a three-dimensional (3D) geometry for the shape based on the 2D position map.

The invention discloses a lane line detection method based on SDAMamba and FFCDConv, and relates to the field of lane line detection. The method comprises the following steps: preprocessing a training set lane line image to obtain an updated training set; improving the DAMamba model to obtain an SDAMamba model for extracting multi-scale features of the updated training set image; extracting depth multi-scale features by using a feature pyramid neck network; mapping the depth multi-scale feature to a Hough parameter space by using DHT and selecting a Hough feature corresponding to a lane; jointly inputting the Huffcharacteristics and the depth multi-scale characteristics into an FFCDConv module, predicting the characteristics of different lane instances, and obtaining a predicted lane position map through a lane decoder; and inputting the predicted lane position map into a linear layer to determine a final lane position, and obtaining a lane line detection image. According to the method, the problem of breakage easily occurring in a lane line detection result can be repaired in a targeted manner, and the confusion rate of adjacent lanes is remarkably reduced.

The invention belongs to the technical field of petroleum geological exploration, and relates to a logging lithology intelligent identification method and system based on well network position map structure learning and feature fusion, and the method comprises the steps: obtaining the original data of logging in a target work area, and carrying out the standardization preprocessing, the original data comprises logging data, well position coordinate data and lithology label data; according to the original data after standardization preprocessing, a well network position map structure is established; performing feature extraction on the well pattern position map structure by using a map neural network to obtain high-dimensional features; fusing the high-dimensional features with original logging features, and obtaining an enhanced feature set to train a lithology classification model; and utilizing the trained lithology classification model to predict the lithology of a to-be-identified well section, and outputting a lithology identification result of the to-be-identified well section. The lithology identification accuracy and the geological rationality can be obviously improved.

The invention relates to the technical field of agricultural automation, and discloses a seed leakage real-time detection method and system for an automatic sugarcane planter, and the method comprises the steps: collecting the distance measurement data and position signals of each seed metering channel in real time through a multi-channel sensor, and obtaining an initial distance measurement data set; calculating the relative distance difference between the channels based on the initial ranging data set, and determining the zero drift distance; when the zero drift distance exceeds a preset threshold value, datum point correction is carried out on the target calibration channel, and calibrated ranging data are obtained by fusing adjacent channel data; performing cross validation on the calibrated data, and judging an accumulative error level; channel cooperation parameters are adjusted according to the accumulative error level, distance measurement values passing verification are integrated through a data fusion method, and an optimized relative distance matrix is obtained; and mapping seed leakage point coordinates through the matrix, generating a position map, extracting reseeding target points, and determining a final seed leakage positioning result. The accuracy of seed leakage detection is improved.

The application relates to an interpretable pedestrian prediction method introducing expert experience in a complex interactive environment, comprising the following steps: acquiring pedestrian information in an environment, wherein the pedestrian information comprises relative position, speed and acceleration information; determining expert experience coefficients of relative position relationships, relative speed relationships and relative acceleration relationships based on the pedestrian information; constructing interpretable position graphs, speed graphs and acceleration graphs based on the expert experience coefficients and the pedestrian information; performing topological structure coding on the position graphs, the speed graphs and the acceleration graphs based on a multilayer graph convolutional neural network to obtain hidden states containing spatial topological information; and performing space-time coding on the hidden states containing the spatial topological information, decoding by using a decoder to obtain predicted pedestrian trajectories. Compared with the prior art, the application has the advantages of fine consideration of interactions between pedestrians, accurate pedestrian trajectory prediction in a complex interactive environment and strong interpretability.

A self-organizing mesh network system is disclosed, comprising: at least one master node generating a timing reference signal; a plurality of regular nodes deriving time synchronization from the timing reference signal; and a wireless medium for communicating location and timestamp information among the plurality of regular nodes, The plurality of regular nodes may be configured to each use communicated location and timestamp information of nearby nodes to independently generate a location map of the nearby nodes. The plurality of regular nodes may be configured to accept an additional regular node. The plurality of regular nodes may be configured to allow a node of the plurality of regular nodes to exit the plurality of regular nodes. The plurality of regular nodes may each use communicated location and timestamp information of the plurality of regular nodes to independently generate a location map of each of the plurality of regular nodes.

Methods and systems for determining a signal-to-noise metric for locations of interest on a specimen are provided. One or more statistics of non-defect signals from background patch images in a test image that are similar to a patch image of a location of interest in the test image are determined. The background patch images are found by searching a reference image for patch images that are similar to the location of interest patch image and finding the corresponding patch images in the test image. The signal of the location of interest in the test image and the one or more statistics are used to determine a signal-to-noise metric for the location of interest. The signal-to-noise metric can be used in applications such as defect annotation, deep learning (DL) model tunability, DL model repeatability, and novel defect detection.

The present application relates to inland waterway inspection method, system and storage medium, including navigation mark detection and water level detection, navigation mark detection includes: step 11. Data acquisition;Step 12. Navigation mark identification;Step 13. Navigation mark positioning;Step 14. The real coordinates of navigation mark are inquired in the preset navigation mark data table, if the real coordinates of navigation mark have corresponding navigation mark in the preset navigation mark data table, then the current position information of the navigation mark is recorded;Step 15. Corresponding detection report is generated;Water level detection includes: step 21. Image acquisition;Step 22. Obtain the water gauge position image;Step 23. The water gauge position image is filtered, smoothed and denoised;Step 24. The water gauge scale and the image information with similar RGB values are extracted by setting threshold, and the scale is extracted by character extraction method, and the final water level information is obtained.The present application can automatically judge whether the navigation mark is damaged, the position is deviated, whether there is missing condition and water level.

The invention discloses a rapid infrared small target detection method and system based on local gray feature discrimination. The method comprises the following steps: receiving infrared image data; gray scale division is carried out on the original gray scale space of the infrared image, and a gray scale histogram is constructed; performing window sliding on the gray histogram to obtain a gray histogram statistical result of a sliding window; obtaining a Gaussian filtering result of the sliding window, completing a target enhancement operation, and obtaining a Gaussian core enhancement value; on the basis of a gray level histogram statistical result, carrying out real-time discrimination on a central pixel of the sliding window, obtaining a target prior response value, and obtaining a target prior mask position map; performing multi-scale contrast calculation on the infrared image to obtain local contrast response values of all potential target points; and taking the pixel position corresponding to the maximum value of the local contrast response value as a target point coordinate, and outputting a detection result. According to the method, the target area can be accurately positioned, and efficient and real-time detection requirements are met.

The invention discloses an image rendering method, device and equipment for a relightable dynamic digital human and a readable storage medium. The image rendering method comprises the steps of calculating a parameterized model corresponding to a current driving posture of a human body and local three-dimensional Gaussian position attributes corresponding to three-dimensional gausses respectively; generating a front two-dimensional position map and a back two-dimensional position map corresponding to the parameterized model by using the target prediction model; predicting and generating a front three-dimensional Gaussian increment position offset attribute graph, a back three-dimensional Gaussian increment position offset attribute graph, a front diffuse reflection map, a back diffuse reflection map, a front roughness map and a back roughness map; obtaining a current sampling light, and calculating the visibility of each three-dimensional Gaussian in the current sampling light based on a point ray tracing algorithm; colors corresponding to the three-dimensional gausses are calculated; and performing human body image rendering according to the current viewpoint and the colors corresponding to the three-dimensional gausses. According to the invention, the applicability of the digital human model to the application scene is enhanced.

The invention discloses a VR image generation method, which comprises the following steps of dividing VR image data into model data and map texture data, respectively storing the model data and the map texture data in a database, and establishing an index table of the model data and the map texture data; extracting an initial position of the VR equipment, retrieving corresponding model data and mapping texture data in the index table according to the initial position, and combining the corresponding model data and mapping texture data to obtain a VR image of the initial position; the method comprises the following steps: performing model data pre-loading on an image area adjacent to a VR image at an initial position, then tracking a moving process of VR equipment, performing multi-stage mapping processing on model data which is not subjected to mapping processing in a real-time VR image, and performing model data pre-loading on an image area adjacent to the real-time VR image at the same time. The defects in the prior art can be overcome, and the generation speed of the VR image is increased.

An autonomous warehouse monitoring robot 100 comprises a propulsion system 106, 108, 110, a processing unit 112 and a scanner array 114 that includes an RFID scanner arranged in a moveable tower, wherein the scanner arrangement is arranged to scan RFID data from at least one package that includes multiple articles. The processing unit is configured to receive RFID scanner data from the scanner array wherein the scanned RFID data relates a horizontal location (P1, P2, fig 1B) for the article in the warehouse, a vertical location (L1, L2, Fig 1B) for the article in the warehouse, and environmental data such as temperature and humidity (148, fig 1C) scanned at the horizontal location and the vertical location for the article. A storage system and method is also disclosed.

To track probes during intracardiac procedures.SOLUTION: A method includes: computing a position of an intrabody probe, which includes one or more electrodes and an electromagnetic sensor, within a heart of a subject, based on an induced signal received from the electromagnetic sensor; ascertaining a set of properties of signals passed between the electrodes and multiple reference electrodes located at respective reference positions; based on the set of properties, deriving an estimated position of the probe from a position map that maps multiple sets of properties to respective estimated positions; in response to a distance between the computed position and the estimated position being greater than a predefined threshold, ascertaining whether an electrocardiographic signal from the subject is saturated; and, in response to the electrocardiographic signal not being saturated, updating the position map so as to map the set of properties to the computed position. Other embodiments are also described.SELECTED DRAWING: Figure 1

A system and method of managing configurations of rearranged servers in a data processing system (DPS). The method includes acquiring infrastructure data associated with a DPS including a first group of servers being monitored by a first group of sensors that generate a first set of sensor data and a second group of servers being monitored by a second group of sensors that generate a second set of sensor data. The method includes querying the DPS to acquire a plurality of network addresses associated with the first group of servers and the second group of servers. The method includes identifying a plurality of DPS locations of the first group of servers and the second group of servers. The method includes generating, based on the plurality of DPS locations, a server location map of the first group of servers and the second group of servers in the DPS.

This application relates to an arbitrary resolution fusion method for hyperspectral images based on a state-space model, belonging to the field of image processing technology. Addressing the problems of poor multi-scale adaptability and fusion accuracy in existing super-resolution image fusion and reconstruction methods, this application provides an arbitrary resolution fusion method for hyperspectral images based on a state-space model. This method uses a global spatial-spectral fusion module and a bilateral adaptive upsampling module to fuse a low-resolution hyperspectral image with a panchromatic image to obtain fusion features. Based on the fusion features and the location map, an arbitrary-scale super-resolution image is obtained. This invention significantly improves the fusion imaging performance of low-resolution hyperspectral images and panchromatic images, achieving high-precision fusion reconstruction of high-resolution hyperspectral images of arbitrary resolution, effectively solving the bottleneck problems of multi-scale adaptability and fusion accuracy in traditional methods.

A golf assist system (500) assists a user on a green. A course map database (251) has stored therein green surface information (511), green speed information (512), and reference point position information (513) indicating a position of a feature reference point on a reference feature. An image-taking unit takes a green taken image (51) including the feature reference point, a cup provided in the green, and a golf ball on the green by a camera. A position calculating unit (210) calculates a ball position (62) by using the green taken image (51), the green surface information (511), and the reference point position information (513). A line calculating unit (220) calculates a putting line (22) from the ball position (62) to a cup position (61) by using the green surface information (511) and the green speed information (512).

The application discloses a skull gender identification method based on multi-modal, and relates to the field of skull gender identification. The method comprises the following steps: reconstructing a three-dimensional skull surface model based on skull CT scanning data to obtain a three-dimensional skull data set; for each three-dimensional skull surface model, adopting Koebe iteration to realize conformal mapping from a three-dimensional multi-connected region to a plane circular region to obtain a two-dimensional unit circular disc, and respectively performing normal calculation and coloring, vertex coordinate calculation and coloring, curvature calculation and coloring to obtain a normal map, a spatial position map and a curvature map; inputting the normal map, the spatial position map, the curvature map and a gender label corresponding to the three-dimensional skull surface model into a multi-modal skull gender identification network to train a skull gender identification model; and when the skull gender identification model receives a normal map, a spatial position map and a curvature map of a target three-dimensional skull surface model, the skull gender identification model outputs a corresponding gender. The application can improve the accuracy, efficiency and generalization ability of identification.

This invention relates to an adaptive method and system for simulating typhoon wind fields by enhancing the influence of complex terrain and the linkage effect between stations. The method involves acquiring meteorological observation station data, low-resolution typhoon wind field data, and high-resolution terrain data through a data acquisition and standardization preprocessing module, followed by standardization processing. Based on the observation station data, a spatial location map and a feature correlation map are constructed. A multi-scale wind-terrain feature fusion module extracts neighborhood wind field and terrain information. A spatial linkage network prediction module combines the fused features with the observed values ​​for prediction. A typhoon wind field model training module optimizes parameters through backpropagation, minimizing the L2 loss between predicted and actual noise. In the typhoon wind field reconstruction module, target grid points are inserted into the station map, edges are constructed, and wind components are calculated to reconstruct the typhoon wind field. This method addresses the problem of insufficient representation of complex terrain and the linkage effect between stations in existing typhoon wind field simulation methods.

A platform receives an input document from a user device and automatically determines a semantic signature for the input document based on a probabilistic distribution of rare words within the input document. The platform automatically scrapes at least one Internet database for additional documents and webpages, determining semantic signatures for each document or webpage. Based on similarity of semantic signatures, the platform automatically constructs and displays a graphical network of documents, wherein each document is represented as a node and similarity of semantic signatures is used to determine the locations of edges between nodes. The graph automatically groups nodes by communities and selects nodes in different communities to promote serendipity of results.

The application discloses a feature map upsampling method and application, and belongs to the technical field of image processing, and comprises the following steps: S1, mapping an input feature map into a position bias map; wherein, elements (Delta x i , Delta y j ) of the i-th row and the j-th column of the position bias map represent position bias at a position ([i / s], [j / s]) of the input feature map; s is a scaling ratio; H and W are respectively the length and the width of the input feature map; S2, adding the elements (Delta x i , Delta y j ) of the i-th row and the j-th column of the position bias map to the position ([i / s], [j / s]) to obtain a sampling position map of the input feature map; S3, performing interpolation on the input feature map based on the sampling position map to obtain a high-resolution feature map, so that the upsampling operation of the input feature map is realized. The application can realize the upsampling with low complexity and fast calculation speed on the basis of ensuring the upsampling performance.

The application provides a building crack detection method based on a neural network, comprising the following steps: S100, acquiring a building crack image and performing illumination compensation on the image; S200, extracting multi-scale features of different resolutions of the building crack in the image after the illumination compensation and fusing the features; S300, determining a crack position by obtaining a crack thermal map through the fused features, focusing on a crack region by using a spatial attention mechanism based on a crack position map and the fused feature map, and detecting a crack width, a crack length and a crack direction.

The invention discloses an underwater vehicle rudderstock position map design method based on CFD numerical simulation, and the method comprises the steps: dividing the chord length of a tail rudder into a plurality of equidistant rudderstock position points, and achieving the systematic expression of rudderstock arrangement through the combination of a balance ratio; on the basis of a CFD (computational fluid dynamics) method, an overlapping grid technology is adopted, a simulation environment under the rudder angle change is constructed, and key performance indexes under different rudder stock position and rudder angle combinations are calculated; extracting feature data through a plurality of groups of simulation results, and drawing a map relationship among the rudderstock position, the rudder angle and the hydrodynamic response; and determining an optimal rudderstock position by using the map. By coupling CFD simulation and structural modeling, parameterization, quantification and visualization of rudder system design are achieved, and the design decision efficiency under multiple working conditions and multiple parameters is improved.