623 results about "Projection plane" patented technology

The invention discloses an orthoimage generation method and device based on three-dimensional Gaussian splashing and a storage medium, and the method comprises the steps: responding to an orthoimage generation signal of a detected building region, obtaining a multi-view image obtained by photographing the detected building region from a plurality of view angles, determining an image acquisition parameter of an image acquisition device corresponding to the multi-view image; performing cross-image matching on the feature points in each image, and determining feature point pairs in each image based on a matching result; based on the feature point pairs and the image acquisition parameters, performing three-dimensional reconstruction on area points in the detected building area represented by the feature point pairs to obtain three-dimensional point clouds, and performing Gaussian distribution processing on the three-dimensional point clouds to obtain a plurality of three-dimensional Gaussian ellipsoids; and determining a two-dimensional projection plane, determining depth information of each three-dimensional Gaussian ellipsoid from the two-dimensional projection plane, and splashing each three-dimensional Gaussian ellipsoid into the two-dimensional projection plane based on the depth information to obtain an orthoimage of the measured building area.

The invention discloses a pipe online defect detection method and system, and relates to the technical field of image detection. The detection method comprises the following steps: placing a to-be-detected pipe in a dark box environment, irradiating the to-be-detected pipe in a rotating state through a light source, displaying a pipe projection on a projection surface, and setting a detection line at the upper end of the projection surface; comparing the detection line with a set reference line in real time, and marking a suspected warping area according to the local offset; performing three-dimensional image reconstruction on the suspected warping area to generate a simulation three-dimensional point cloud model; the method is technically characterized in that preliminary judgment passes through a grading strategy of'rapid screening-precise reinspection ', and the detection speed and precision are balanced; by means of the optical path principle and threshold value comparison, positioning operation of the whole-surface suspected warping area can be rapidly completed, only data of a local area (suspected warping area operation) is analyzed, the hardware cost and energy consumption are greatly reduced, and the high-speed online detection requirement is met.

The invention provides a tunnel quality safety detection system and method based on laser radar data, and the method achieves the fusion and complementation of crack detection based on point cloud density change and crack detection based on curvature change, and achieves the quick recognition and precise detection of a crack region. The potential crack area is rapidly positioned through point cloud density change detection, and the boundary of the crack is further accurately positioned in combination with curvature change detection, so that the accuracy and integrity of crack detection are improved. For tiny deformation which is very tiny in change in point cloud density and curvature and is difficult to directly detect, a projection plane is further determined on the basis of the identified crack, and after point cloud data is projected to a two-dimensional plane, the characteristics of the tiny crack in the image are highlighted through image enhancement, so that the crack is identified. And then an edge detection algorithm is used to identify the micro crack so as to make up for the defects of a previous point cloud density and curvature detection method in the aspect of micro crack detection, and the accuracy and integrity of crack detection are further improved.

The invention discloses a multi-person point cloud human body posture reconstruction method and system based on a millimeter wave radar, and the method comprises the steps: firstly generating a three-dimensional point cloud through radar echo preprocessing, and achieving the noise filtering and target separation through the combination of detection region division and a multi-stage DBSCAN clustering algorithm; secondly, a multi-scale feature fusion neural network is constructed, projection plane interference is isolated through grouping convolution, space details are reserved based on a deep high-resolution module, key feature extraction is enhanced through an improved ResNet18 embedded coordinate attention and channel attention mechanism, and end-to-end human body 23 joint three-dimensional coordinate regression is achieved. The system supports synchronous reconstruction of postures of multiple persons through real-time point cloud classification and parallel reasoning, and the frame rate reaches 30-40 FPS in a complex scene. The method gives consideration to privacy protection and high robustness, and can be widely applied to the fields of smart home, smart health care and security monitoring.

The embodiment of the invention provides a target segmentation method and device for a multi-view image, electronic equipment and a storage medium, and the method comprises the steps: obtaining scene images of different views obtained by a plurality of image collection devices which are deployed in a target scene in advance, and carrying out the three-dimensional Gaussian reconstruction, and obtaining a three-dimensional Gaussian point cloud corresponding to the target scene; in response to a segmentation prompt instruction corresponding to a target segmentation article in the scene articles, performing two-dimensional article segmentation on the scene image, and generating a corresponding binary segmentation mask; carrying out projection processing on the three-dimensional Gaussian point cloud to obtain projection planes of a plurality of visual angles, and carrying out Gaussian point filtering on each projection plane based on the binarization segmentation mask to obtain a filtered three-dimensional Gaussian point set; and based on the pose point cloud estimation information, the filtered three-dimensional Gaussian point set is sequentially splashed to device planes corresponding to a plurality of image acquisition devices to obtain a corresponding target segmentation result, so that the accuracy of object two-dimensional image segmentation in a target scene is greatly improved.

The invention provides a BIM model three-dimensional projection plane extraction method, electronic equipment and a medium, and the method comprises the steps: carrying out the triangularization processing of a BIM model, and obtaining a triangularization mesh model; calculating the projection area and the normal direction of the triangle in the P direction, and deleting the triangle of which the vertex is repeated and the projection area is 0 or the normal direction is opposite to the P direction; calculating a value range of coordinate values of all vertexes of the remaining triangles in the P direction, and dividing the value range into a plurality of intervals; classifying the remaining triangles into a plurality of intervals to obtain a plurality of triangle sets; for each triangle set, extracting a two-dimensional vector surface; and sorting the two-dimensional vector surfaces according to a sequence from small projection height to large projection height, and extracting to obtain a three-dimensional vector surface. According to the method, projection plane extraction is performed after triangularization based on the BIM model, the influence of a professional design software production environment and a multi-source heterogeneous storage format is eliminated, and universality is achieved; compared with two-dimensional expression, the method for extracting the three-dimensional projection surface result provides finer geometric range description, and calculation is simple.

The invention relates to a region-limited virtual touch interaction method and related equipment, and the method comprises the following steps: carrying out the boundary scanning of a region-limited projection plane, and obtaining a projection region contour coordinate set; performing region division on a projection plane based on the projection region contour coordinate set to obtain interaction region grid data; performing three-dimensional feature point collection on the hand of the user through a depth camera to obtain a hand key point space coordinate sequence; when the hand key point space coordinate sequence is within a depth interaction threshold range of interaction area grid data, collision detection is carried out on the hand key point space coordinate sequence based on the interaction area grid data, and a touch event judgment result is obtained; and performing visual feedback rendering on the interaction area grid data based on the touch event judgment result to obtain interaction state display data, thereby solving the technical problem of low interaction precision caused by difficulty in accurately positioning a gesture of a user in an open and irregular space in an existing solution.

The invention discloses a PHC tubular pile perpendicularity measuring method based on double-camera equipment with an IMU module, and the method comprises the steps: horizontally placing two cameras on the ground xOy, and adjusting the two cameras to enable the optical axes Ay and Bx of the two cameras to be perpendicular to each other; respectively reading the roll angles alpha 1 and alpha 2 of the two cameras by using an IMU module; the positions of the target pile body OT in the pictures shot by the two cameras are recognized respectively, and the inclination angles beta 1 and beta 2 of the target pile body OT in the pictures shot by the two cameras are obtained through calculation; the inclination angles theta 1 and theta 2 of the target pile body OT in the orthogonal projection planes SA and SB under the ground coordinate system Oxyz are obtained through the angle relation, and the perpendicularity of the pile body OT is obtained through calculation. Multi-view stereoscopic vision and dynamic structure analysis are adopted, real-time recognition and positioning of the target pile body are achieved, the camera coordinate system is calibrated in combination with the IMU, and the actual construction precision requirement of the pile foundation is effectively met.

The application relates to the technical field of terminals, and discloses an electronic device and a cover. The electronic device comprises a cover, a display screen assembly and a main body. The cover is used for shielding a gap between the display screen assembly and the main body. The cover comprises a first part and a second part which are connected. The first part covers at least part of the edge of the display screen assembly. The second part extends from the first part towards the main body. The inside of the main body comprises an acoustic cavity, and the cover and / or the main body is provided with an acoustic outlet channel which is connected to the acoustic cavity and the outside of the mobile phone. The acoustic outlet of the acoustic outlet channel has an area greater than 0 in the region of the orthographic projection of the side projection plane. The side projection plane is perpendicular to the plane in which the display screen assembly is located and parallel to the edge of the display screen assembly corresponding to the acoustic outlet channel. The electronic device described above cannot be observed from the front, the acoustic outlet channel and the internal components cannot be observed, the appearance is improved, meanwhile, the size space of the display screen assembly can be increased, and the space utilization rate of the electronic device is improved.

The invention relates to the technical field of intelligent traffic, in particular to an intelligent vehicle lamp system and a control method thereof.Optical sensing equipment is used for collecting environment information around a vehicle, a pre-projection image is determined according to the environment information, optimal projection parameters are calculated according to the reflectivity and flatness of a projection plane, and the parameters are sent to a light module; projecting according to the calculated focal length and angle parameters; and adaptively adjusting the frequency of detecting the environment information by the optical sensing equipment by using a cyclic decision detection algorithm to realize adaptive control on different roads, weather and vehicle states. According to the method, the light brightness can be automatically adjusted, unreasonable projection areas can be avoided, safety prompt projection is carried out according to road conditions, the accuracy and safety of vehicle illumination are improved, and the risk of traffic accidents is reduced.

At least one embodiment relates to a method comprising obtaining an occupancy information indicating if blocks of a 2D regular grid defined over a projection plane are associated with depth information of an orthogonal projection of at least one point of a point cloud onto said projection plane; embedding said occupancy information in an image storing said depth information; and encoding said image.

The invention relates to the technical field of intelligent vehicle lamps, and particularly provides an intelligent vehicle lamp control method based on color avoidance and depth correction, and the method comprises the steps: determining a corresponding heterochromatic region on a pre-projection image according to the color deviation tensor of a projection plane and a global color fitting plane, and controlling a vehicle lamp to carry out the color avoidance projection of the heterochromatic region, according to the depth difference tensor of the projection plane and the ideal depth plane, determining a corresponding depth change area on the pre-projection image, and controlling the vehicle lamp to change the projection size to realize depth correction, so as to ensure that the finally projected image is not interfered by the color of the actual road surface; and discontinuous and unclear projection images in the visual field of the driver due to road surface potholes are avoided. According to the invention, the light brightness and the local projection size of the vehicle lamp can be adjusted according to the collected information, so that the projection of the vehicle lamp is not interfered by the road surface color and the road surface unevenness.

The invention relates to a projection fusion display control method, which comprehensively monitors ambient light through a light sensor and a camera, analyzes light intensity, color and direction, and calculates an influence value of light on a projection effect so as to accurately adjust the brightness and color of a projection portrait. And secondly, detecting the flatness of the projection surface by using a high-precision laser range finder, generating a flatness influence value, correcting the image, and compensating the effect influence caused by the unevenness of the projection surface. In addition, the technical scheme also relates to the identification of the projection plane image to obtain the display template, the regional picture correction of the image is carried out according to the template and the display parameters, and finally the optimal image display frame is generated through fusion for projection. According to the scheme, the definition and the color rendition degree of the projection picture are improved, the stability and the consistency of the projection effect are enhanced, and better projection experience is brought to a user. By finely processing the ambient light and the flatness of the projection surface, the technical scheme effectively improves the overall application effect of the projection technology.

The invention relates to the technical field of deep neural network training, and discloses a deep neural network training diagnosis method and system based on gradient visualization and loss curved surface analysis, and the method comprises the steps: obtaining a parameter snapshot of a neural network trained to t-th iteration, sampling k random detection vectors, and calculating a Hessian vector product; performing Lanczos iteration on each random detection vector to obtain k symmetric tridiagonal matrixes and calculating a characteristic value density spectrum; mapping the k symmetric tridiagonal matrixes into orthogonal projection base vectors, constructing a two-dimensional projection plane, determining a saddle point state according to the integral of the characteristic value density spectrum in a negative value region, determining a flat region according to the peak value of the characteristic value density spectrum in a zero value region, and determining a sharp minimum value according to the peak value of the characteristic value density spectrum in a positive value region; according to the method, the descending path with the steepest curvature can be captured, and the dominant mode of the dynamics can be optimized, so that the projected loss curved surface truly reflects the geometric structure change in the training process, and the pathological diagnosis of the training process based on the second-order geometric information is realized.

The present invention relates to methods for visual display of images of real shelvings with products for analysis of shelving contents. There are provided methods for constructing a shelving front panorama. A method comprises: capturing color image frames of the shelving, displaying the shelving and its contents; reconstructing a shelving 3D model based of depth data and capturing device position data for each frame; determining a projection plane corresponding to the shelving front edge; stitching the color image frames of the shelving by projective transformation of each color image frame of the shelving onto the projection plane. The resulting shelving front panorama displays the shelving and its contents. The disclosure eliminates the need for observing frontal orientation of each shelving image frame when capturing images of the shelving; the capturing can be made along an arbitrary path.

The embodiment of the utility model provides a shaping device and a battery device production line. The shaping device comprises a first shaping piece and a second shaping piece which are arranged in the first direction, the first shaping piece comprises a first surface and a second surface, the first surface is used for containing the adapter, the second surface is located on the side of the first surface in the second direction, and the first surface and the second surface are connected to form a first connecting position. The first connecting part is used for shaping the first tab. The second shaping piece comprises a third surface and a fourth surface, the first surface and the third surface are oppositely arranged, the fourth surface is located on the side of the third surface in the second direction, the third surface and the fourth surface are connected to form a second connecting position, and the second connecting position is used for shaping the second tab. And on a projection plane perpendicular to the first direction, the orthographic projection of the second connecting part is closer to the center of the second shaping part than the orthographic projection of the first connecting part. According to the technical scheme provided by the invention, the reliability of the battery device can be improved.

The invention provides a humidifying tank and breathing assisting equipment, the humidifying tank comprises a tank body, a base plate and a sealing piece, the sealing piece is installed between the inner surface of the side peripheral wall of the tank body and the base plate, and the sealing piece is used for keeping the tank body and the base plate sealed and connected; a first fixing part is arranged at the open end of the tank body, the chassis comprises a second fixing part, and the second fixing part is connected with the first fixing part in a matched mode, so that the chassis is installed at the open end of the tank body and defines an inner cavity with the tank body to contain liquid; a limiting part is arranged on the tank body or the base plate and used for being matched with a main machine of the breathing assisting device to install the humidifying tank. The orthographic projection of the limiting part and the orthographic projection of the sealing piece on the same projection plane at least partially coincide in the projection direction perpendicular to the axial direction of the side peripheral wall. The technical problem that the sealing ring shields the water level height in the humidifying tank is mainly solved, and low-water-level recognition of the breathing assisting device is achieved; the chassis and the tank body are respectively provided with a fixing part, so that firm connection is ensured, and meanwhile, disassembly and assembly are convenient.

The first aspect of this application proposes a method for fusing laser point cloud and visible light image, including: S1, performing spatial dimensionality reduction on the point cloud data to obtain two-dimensional point cloud data; S2, extracting feature points of the two-dimensional point cloud data and feature points of the visible light image data through the Scale-Invariant Feature Transform (SIFT) algorithm; S3, determining the projection plane, projecting the point cloud data through the projection model to generate a depth-echo intensity image, and constructing a 3D-2D coordinate mapping relationship between the point cloud data and the depth-echo intensity image; S4, mapping the feature points of the visible light image data to the point cloud data according to the 3D-2D coordinate mapping relationship. This solution can achieve a good image registration and fusion effect, can be applied to various platforms, and has a wide range of applicable scenarios. The second aspect of this application proposes a system for fusing laser point cloud and visible light image.

The invention relates to the technical field of tooth root angle measurement, and provides a tooth root angle three-dimensional intelligent measurement method based on an oral cavity CBCT image, and the method comprises the steps: constructing a Teeth-MambaSeg tooth segmentation neural network model for carrying out tooth segmentation on a to-be-measured oral cavity CBCT image; a training database is constructed, and the Teeth-MambaSeg tooth segmentation neural network model is trained; performing tooth segmentation on a to-be-measured oral cavity CBCT image through the trained Teeth-MambaSeg tooth segmentation neural network model, and projecting a single segmented tooth to a simulated two-dimensional projection plane to obtain a tooth simulated projection image; according to the tooth simulation projection image, feature points related to tooth root angle calculation are extracted, and three-dimensional space coordinates of the feature points are obtained; and obtaining a tooth root angle three-dimensional measurement value according to the three-dimensional space coordinates of the feature points. According to the invention, personal errors of measurement point positioning can be effectively reduced, and the stability and accuracy of tooth root angle measurement are improved.

A remote operator terminal is used by a remote operator for a remote operation of a moving body. A first image is captured at a first timing by a camera mounted on the moving body from a first point of view. A second point of view is one at a second timing later than the first timing. A homography process converts the first image viewed from the first point of view into a second image viewed from the second point of view based on perspective projection transformation using a projection plane associated with the camera. The remote operator terminal acquires a combination of a plurality of second images by applying the homography process using a common projection plane to a combination of a plurality of first images, and displays the combination of the plurality of second images on a display.

Provided in the embodiments of the present application are a projection control method, a projector and an electronic device, which are applied to the technical field of projection imaging. The projector comprises a projection assembly and a camera. The method comprises: on the basis of a first data signal, a projection assembly emitting projection structured light, the first data signal being used for indicating a first image, the projection structured light being used for forming on a projection plane a projected image of the first image, and the first image comprising a user projected image and / or a calibrated feature image; a camera photographing a first area, so as to obtain a second data signal, the second data signal being used for indicating a second image; and, according to the second image, obtaining a first control signal, the first control signal being used for adjusting projection parameters of the projection structured light. In the embodiments of the present application, the projection assembly and the camera of the projector form a structured light apparatus which performs algorithmic processing according to the second image, so as to calculate the projection parameters. The present solution can be applied to projector products of different costs so as to rapidly and automatically adjust projection parameters with high precision.

The present invention generally relates to a technology for recognizing objects in a Around View Monitoring (AVM) video for a vehicle. In particular, the present invention relates to a DNN-based object recognition method for multi-channel fisheye images of a vehicle AVM apparatus, by projecting multi-channel unit images of fisheye-lens cameras of the vehicle AVM apparatus onto a cylindrical projection plane so as to obtain multi-channel unit projection images, combining the multi-channel unit projection images on a single image so as to obtain a composite projection image, and inputting the composite projection image to a Deep Neural Network (DNN) model so as to obtain object recognition output. The invention has an advantage that AVM apparatus can effectively assist in safe driving by monitoring objects around the vehicle. Further, the present invention has an advantage that object recognition may be effectively achieved by the insufficient computing resources of vehicles.

The invention relates to the technical field of construction quality detection, in particular to a building construction quality automatic detection method and system.The building construction quality automatic detection method comprises the following steps that vertical projection plane coordinates of multiple components in a building fabricated structure are obtained, a horizontal edge vertex coordinate set is selected, and the geometric center of the components is obtained through calculation and averaging; and generating a component positioning center vector group according to the component number sequence connection points. According to the method, through extraction of vertical projection coordinates of the components, construction of a positioning center vector, establishment of sequence expression of the position relation between the components, judgment of the consistency of the offset difference change rate and the included angle, continuous offset trend extraction, extraction of the gravity center of the components in the section, calculation of the deviation degree of a connecting line to a theoretical axis, and reflection of the overall offset trend of the component section. And identifying the system characteristics of the error according to the offset consistency and the amplitude distribution. According to the process, the component position, direction and aggregation offset characteristics are integrated in a serialization, vectorization and gravity center weighting mode, and the capacity of trend recognition and cluster error judgment is improved.

The invention belongs to the technical field of image processing, and provides an image blurring correction method and device, electronic equipment and a medium, and the method comprises the steps: carrying out the projection of an image and a standard map based on a blurred image, building a Gaussian-Kruger projection model, and obtaining auxiliary parameters, calculating to obtain a horizontal coordinate and a vertical coordinate in a projection plane rectangular coordinate system based on the auxiliary parameters, converting the pixel points into coordinates of the plane rectangular coordinate system, marking the converted coordinates of the pixel points as converted coordinates, generating a sampling signal if the converted coordinates and the projection coordinates are inconsistent in the same plane rectangular coordinate system, and outputting the sampling signal. The method comprises the following steps: sampling a sampling signal, re-sampling an image by adopting a bilinear interpolation method based on the sampling signal, marking a pixel point as a deteriorated pixel point if the pixel value of the pixel point of the sampling image is smaller than the pixel value of the pixel point of an original image, obtaining a deteriorated performance value based on the deteriorated pixel point, obtaining error data based on an adjustment signal, and outputting the error data. And adjusting the deteriorated coordinates according to the error data.

The invention relates to the technical field of electronic equipment, in particular to an accessory main body and an accessory, the height of the accessory main body is smaller than the contour diameter of a projection plane, and the projection of the accessory main body on a plane perpendicular to the height direction of the accessory main body is the projection plane. The accessory body is provided with at least one first light-emitting element, and light of the first light-emitting element is emitted from the peripheral side of the accessory body. The height of the accessory main body is set to be smaller than the contour diameter of the projection surface, the accessory main body is of a flat structure, wearing of the accessory main body is facilitated, the accessory main body is provided with at least one first light-emitting element, and light of the first light-emitting element is emitted from the peripheral side of the accessory main body. The accessory body can be tracked and positioned through the light of the first light-emitting element, the positioning accuracy is improved, and the user experience is improved.

An endoscope includes a distal end member that is connected to a distal end side of an insertion part and that is made of a conductive material, and an insulating ring that is mounted on a base end side of the distal end member and that has an insulating property, in which the distal end member has a first region in which an observation window, an illumination window, and a treatment tool outlet port are disposed, and a second region that is provided on a base end side of the first region and on which the insulating ring is mounted, and in a case in which the first region and the second region are projected onto a projection plane perpendicular to a longitudinal axis of the insertion part, a maximum outer diameter of the second region is smaller than a maximum outer diameter of the first region.

The invention discloses an interactive projection system and method based on structured light perception and image anti-distortion, and relates to the technical field of interactive projection, and the method comprises a structured light perception module which is used for obtaining a projection plane depth image and an object image in a projection area; the projection modeling and object recognition module is used for constructing a change relation from a plane coordinate to a three-dimensional world coordinate according to the projection plane depth image, and recognizing object state data according to the object image; the image control and projection display module is used for determining an original image according to the object state data; the image anti-distortion processing module is used for performing pre-transformation processing on the original image based on the grid model and the projection device parameters to generate an anti-distortion image; and the image control and projection display module is also used for projecting the anti-distortion image to the projection area in real time. The interactive experience and the projection effect can be remarkably improved.

The invention relates to the technical field of image processing, and discloses an image processing method of an LED dome screen display screen, comprising the following steps: S1, according to the size of the LED dome screen display screen and the size of a source image supported by a playing system, generating a virtual target image having the same shape as a dome screen, the size of the virtual target image being proportional to the actual size of the dome screen; s2, performing plane cutting on the virtual target image according to each row of pixel points forming the LED dome screen display screen to form corresponding hemispherical arcs; and S3, unfolding the generated hemispheroid arc and projecting the hemispheroid arc into a two-dimensional projection plane. Through accurate image mapping and deformation processing, the visual effect of a source image presented on an LED dome screen is the same as that of the source image presented on a flat panel display screen, an adaptive mapping algorithm is used, the deformation degree of the image can be dynamically adjusted according to image features, it is ensured that each frame of image can keep high-quality display at different visual angles, and the image quality is improved. And the viewing experience of the user is greatly enhanced.

The invention relates to a method for producing an optical device (1) comprising at least one light source (16), at least one light guide (10) having an entry surface (11) facing the light source (16), and having an exit surface (12) formed at an end of the light guide (10) lying opposite the entry surface (11), in order to allow light entering into the light guide (10) to exit same, and a projection optics (20) which is designed to project the light exiting the exit surface (12) onto a projection plane (26) provided at a distance from and at an angle of inclination (γ) relative to a main direction of emission (Z) of the optical device (1). The method comprises the following steps: specifying a target distribution of the illuminance in the image (15) projected into the projection plane (26) by the projection optics (20); determining a non-homogenous distribution of the illuminance (Ev) of the light across the exit surface (12) from the distribution to be specified of the illuminance in the image (15) projected into the projection plane (26); determining a three-dimensional shape of the at least one light guide (10) with which the non-homogenous distribution of the illuminance (Ev) of the light across the exit surface (12) is generated when light is emitted by the at least one light source (16); and producing the light guide (10) with the three-dimensional shape and assembling the components of the optical device (1).

Disclosed is an electronic device including a projection component. The electronic device identifies whether a touch sensing device configured to detachably couple to the electronic device is coupled to the electronic device to sense a touch input of a user on a soft keyboard. Based on identification that the touch sensing device is coupled to the electronic device, the electronic device controls the projection component to project a soft keyboard corresponding to a first keyboard layout on a projection plane.