63results about "3D modelling" patented technology

The invention discloses a method of automatic generation for digital cities, applied to a general-purpose computer system and comprising the following steps: obtaining the remote sensing land image of a presumptive area and monitoring the shadow lengths of all the objects in the remote sensing land image through an algorithm for detecting shadows; vectoring the remote sensing image to obtain the shapes of different objects and match with the position of the shadow to obtain the lengths of the objects; recognizing the different objects in the remote sensing image according to the characteristics of the image in an image database; forming automatically the three dimensional model of the objects in the area according to the types, the shapes of the base, the shapes of the footstock and a highly combined model base of the objects in the area. The method of the invention can form automatically large range digital cities to meet the need of time effectiveness in city application, and only needs remote sensing images and DEM as basic data so that the invention has low cost and the possibility of application in government, commerce, living, etc.

The present invention relates to a gear construction method, including the steps of: i) setting gear parameters and an XYZ coordinate system; ii) according to the gear parameters, creating a first curve associated with an involute of a tooth profile of a target tooth in an XZ plane, and creating a second curve associated with the involute of the tooth profile of the target tooth in a YZ plane; iii) combining the first curve with the second curve to form a third curve and projecting the third curve onto an XY plane to obtain the involute of the target tooth; and iv) constructing a tooth profile of the target tooth and form a solid model of the entire gear by using the involute. The present invention also relates to a digital apparatus associated with the above gear construction method.

The invention relates to a three-dimensional pavement crack image generation system and method. The system includes an information acquisition subsystem, a data processing subsystem and the like, wherein the information acquisition subsystem includes an image acquisition module and a crack depth acquisition module; the image acquisition module includes a shooting control sub-module, a CCD line scan camera and a structured light projector; and the data processing subsystem includes an information storage module, an image registration module, a visible light image crack detection module, a structured light image space texture extraction module and an image fusion module. The method comprises the following steps: step S100, collecting sample crack information, and building a crack depth prediction model; step S200, collecting a target rack image, extracting crack features and space texture, and calculating the target crack depth. Compared with the prior art, the three-dimensional pavement crack image generation system and method provided by the invention are simpler and more feasible and high in accuracy, and have broad application prospects.

The invention discloses a tumor MRI weak supervised learning analysis modeling method, which comprises the following steps: S1, constructing a tumor MRI segmentation network based on a full convolutional neural network to realize tumor MRI preliminary segmentation; S2, taking the tumor MRI segmentation network as a generator, taking the classification model as a discriminator, and proposing a generative adversarial training algorithm of the tumor MRI segmentation network; and S3, taking unlabeled image data in the medical image as additional input condition information of the generator model and the discriminator model, guiding a data generation process, and enhancing the stability of the segmentation-generative adversarial network. The invention also discloses a model established by usingthe tumor MRI weak supervised learning analysis modeling method. According to the method, the tumor target region and the endangered organ can be more accurately and automatically segmented under thecondition of fewer training samples, and the method is applied to target region and endangered organ sketching of tumor radiotherapy.

A method to transmit a label between two images, characterized in that the method includes the following steps:

• • providing a first image, the first image comprising several sets of connected points, each set being characterized by a label,
  • providing a second image,
  • from the second image, determining several sets of connected points,
  • superimposing the two images to determine the common parts and non-common parts of the sets in the first and second image,
  • giving each common part of the second image, the label of the set in the first image with which said part is common,
  • giving each non-common part of the second image in contact with a single set of connected points in the first image, the label of said set,
  • giving a new label to each non-common part of the second image not in contact with any set in the first image.

The invention relates to a medical image processing method which is characterized in that images formed by three-dimensional fast spin echo imaging collected by a magnetic resonance imaging system in the focus area of a patient and blood vessel images collected by a digital subtraction angiography machine are subjected to reconstruction, color inverting, cutting, 3D-3D fusion and 3D-2D registering through a post-processing work station of the digital subtraction angiography machine, so that three-dimensional blood vessel graphs can be overlapped on a real-time radio-fluoroscopic image. Therefore, by adopting the medical image processing method, real-time imaging guide can be performed during nerve involvement surgeries for blocked blood vessels, the near end and far end of the closed section of the blood vessel can be shown simultaneously through imaging, and physicians can clearly get known of conditions in the focus areas and further can perform operating better.

The invention discloses a quick estimation and three-dimensional display method for visual green land areas of city floors. The method comprises the following steps of 1) calling a building layered modeling algorithm to perform base contour extraction and automated three-dimensional modeling on building digital surface model data so as to generate corresponding Shp graphic layer and Multipatch model; 2) calling a floor visual green land estimation algorithm to perform ornamental green land area automatic estimation on extracted building floors; 3) updating attribute fields of a building three-dimensional model and a building base contour model, and performing layered color-setting display. According to the method, a vertical city visual green land analysis concept is innovatively proposed, automated estimation and quick modeling display functions are realized at the same time, and manpower cost of actual measurement is reduced; and the method meets partial demands of people such as city residents, city planners, decision makers, house dealers and the like in related aspects of green landscapes, and has a certain social benefit and economic value.

A 3-D reconstruction method based on Taylor exponent expression surface structured light is provided, The method mainly comprises the following steps: generating and projecting a surface structure grating based on a special exponential expression, acquiring an image, and calculating the phase information of a projector-free gamma error contained in the acquired image based on a corresponding novelphase decomposition method. Thus, through the phase information of no projector gamma error, calculating the three-dimensional measurement result, and eliminating the effect of projector gamma on themeasurement result, and improving the measurement precision.

A method for a computer system includes determining a plurality of illumination modes associated with a plurality of scene descriptors, wherein the plurality of scene descriptors includes a first scene descriptor and a second scene descriptor, determining a first plurality of weights, wherein each weight from the first plurality of weights is associated with an illumination mode from the plurality of illumination modes, determining illumination data associated with the first scene descriptor in response to the first plurality of weights and in response to the plurality of illumination modes, determining a second plurality of weights, wherein each weight from the second plurality of weights is associated with an illumination mode from the plurality of illumination modes, and determining illumination data associated with the second scene descriptor in response to the second plurality of weights and in response to the plurality of illumination modes.

A method for segmenting mesh models in digital images includes providing (11) a mesh surface model (V, F, E) of a shape embedded in a 3-dimensional space, providing (12) a ground truth segmentation of the mesh surface model, constructing (13) a feature vector v_i composed of underlying geometric properties of the mesh at each vertex point, using (14) the ground truth segmentation and the feature vectors to determine weights w_ij for edges e_ij where the weights w_ij are the probabilities of a random walker crossing edges e_ij between vertices v_i and v_j in a random walker segmentation initialized with at least two seeds of different labels.

The invention discloses a method for obtaining a tunnel excavation face surrounding rock point cloud model. The method comprises the following steps: (1) selecting camera parameters according to the tunnel environment; (2) selecting shooting time according to a tunnel construction procedure; (3) selecting a shooting mode, a shooting position and a mark point position according to the excavation length; (4) selecting shooting parameters according to the environment of the tunnel excavation section; and (5) acquiring a surrounding rock image set after tunnel excavation, and processing the imageset to obtain a three-dimensional point cloud model of the surrounding rock after tunnel excavation. Compared with the prior art, the method has the advantages of being rich in collected data size, low in cost, rapid to operate, high in flexibility and the like, a geological engineer can directly shoot and record when observing surrounding rock through geological sketch after excavation, the siteconstruction progress is not affected, shooting personnel can flexibly change the shooting angle and position, and the problem that rock outcrop is mutually shielded is solved as much as possible.

The invention discloses a point cloud acquisition method, device and system and a storage medium. A six-axis robot carries a single line laser three-dimensional camera to move back and forth on the sliding chute, different rows of the whole scanning area are scanned, large-view-field panoramic scanning is realized; structure is very simple, correspondingly, installation and control are very simple, simplification of the structure and control is achieved, and an overlapped area exists between every two adjacent scans, so that after rough splicing is conducted, the rotation matrix and the translation matrix can be calculated only based on the point cloud of the overlapped area, the number of the point clouds is greatly reduced, the fine splicing time is shortened, and the splicing precisionis improved.

The invention discloses a controllable binning method of a virtual maize leaf model. The method belongs to the technical field of virtual agriculture and includes the following steps: collecting the surface feature points of maize leaves; establishing a virtual maize leaf model based on NURBS curved surface by taking the feature points as offset points; determining the shape feature key points ofthe maize leaf model; and according to the feature key points, using the inverse operation of a multi-line section splitting algorithm to conduct the mesh simplification of the curved surface. The method can maintain the key shape of leaves during the mesh simplification and binning process and realize the multilayer and multiscale mesh simplification and binning of the virtual maize leaf model, thus being capable of dynamically adjusting the balance of computational accuracy and computational efficiency.

A virtual debugging system for intelligent manufacture is mainly composed of industrial control equipment, a PC client, a Redis database module and a simulation module, wherein the industrial controlequipment comprises an industrial robot system, a PLC, a CNC lathe and a machining center controller; the PC client comprises data acquisition module and an interactive interface module; the PC clientmakes each equipment driver into a plug-in based on the OSGI.NET plug-in framework, collects the signals and data of the industrial control equipment in real time, and storages the data in the Redisdatabase; the PLC and NC equipment perform signal interaction through the database; the simulation module reads the data in the database through the Python program to drive the robot and machine modelfor processing; at the same time, the status information of the model is also stored in the Redis database; and the PC client reads the status information of the model in the Redis database in real time and writes the same to the PLC, so that the conditions in the PLC are met to schedule the robot, CNC lathe and machining center to perform tasks to achieve the fusion of reality and reality, whichcan verify the feasibility of intelligent manufacture solutions.

The embodiment of the invention discloses a generation method and device for a virtual reality surgery scene. The method comprises the following steps that: collecting the sign data of a target human body, and generating a virtual reality human shape model according to the sign data, wherein the sign data comprises three-dimensional data, the index data of each human body position, the index data of each human body tissue and the like; when a surgery operation instruction input by aiming at the virtual reality human shape model is received, obtaining the surgery operation information, wherein the surgery operation information comprises a surgery operation gesture; and carrying out synthetic processing on the virtual reality human shape model and the surgery operation information to generate a virtual reality surgery scene, and outputting the virtual reality surgery scene. When the method is adopted, the surgery is trained to enrich surgery operation ways and improve the success rate of the surgery.

The invention provides a skull correction information acquisition method, a storage medium and electronic equipment, and the method comprises the steps: determining a foresight datum line and a foresight datum plane through a forehead foresight point and a chin foresight point in an initial side view, determining a skull central axis vertical line through the foresight datum plane, and building athree-dimensional coordinate system through employing the skull central axis vertical line as a reference; obtaining asymmetric information of the skull by utilizing three-dimensional coordinates of the left anatomical mark point and the right anatomical mark point which correspond to each other in anatomy on a three-dimensional coordinate system; using a principle that two points determine a straight line, and combining and analyzing at least two groups of asymmetric information to obtain a force application point, an application force and a pressure application direction required for correcting the skull to mirror symmetry, so that medical personnel or researchers can research a lesion source and a treatment scheme according to the skull correction information.

The invention relates to the field of automation, and discloses a garment showing method and device, so as to solve the technical problem in the prior art that provided image information of network garments is not accurate enough. The method includes: determining a central point of a first garment piece which corresponds to a first body structure of a human body and is any one of M garment pieces contained by a garment, M being an integer greater than or equal to 1; generating a cylinder which surrounds the first body structure of a human body model; performing operation on the central point of the first garment piece and a first matrix, and then moving the central point from a first position to a second position; and flatly laying the first garment piece to the cylinder, and then establishing a three-dimensional garment model of the garment. The technical effect of improving accuracy of image information of network garments is achieved.

The invention discloses an image processing method and device, a storage medium and electronic equipment. A computer vision technology is introduced into the method. The method comprises the steps that a three-dimensional feature volume is constructed according to image features of a target image and a plurality of adjacent images and positions of pixel points in the target image, and the three-dimensional feature volume is formed by stacking a plurality of cost matching images in the depth hypothesis direction; according to the depth hypothesis direction, the probability that pixel points inthe target image are mapped on different depth hypotheses is calculated in sequence, and the probability that the pixel points are located on the current layer depth hypothesis is determined based onthe cost matching graph and the historical cost matching graph on the current layer depth hypothesis; and a depth map is generated according to the probabilities of the pixel points in the target image on different depth hypotheses. According to the scheme, the high-resolution image can be processed to obtain the high-resolution depth map, and the accuracy of the constructed depth map is ensured while the memory occupation is reduced.