41 results about "Monocular" patented technology

The invention discloses a monocular and binocular cooperative localization and mapping method and device for underwater refraction compensation, and the method comprises the steps: extracting checkerboard angular points, compensating an underwater light path through employing a two-time vector correction model based on a Snell law, and precisely calibrating the internal and external parameters of a binocular camera. According to the method, an underwater real light path is effectively restored, and calibration errors caused by medium refraction are remarkably reduced. Then epipolar correction is performed on the corrected left and right images, so that the geometric consistency of stereo matching is ensured; and finally, monocular and binocular poses are fused, and seven-degree-of-freedom similarity transformation global alignment is applied, so that the problem of monocular scale uncertainty is effectively solved, and high-precision and high-robustness three-dimensional mapping is realized by means of binocular real scale introduction. Key links such as camera pre-calibration, refraction compensation, image correction, monocular tracking, binocular depth measurement, scale alignment and the like are connected in series to form an integrated process, so that the underwater vision positioning and mapping precision is improved.

The invention discloses an endoscope monocular dynamic scene reconstruction method based on dynamic Gaussian splashing and motion tracking, and the method comprises the steps: firstly obtaining a dynamic scene image, obtaining an initial three-dimensional track through introducing auxiliary prior information, and then obtaining three-dimensional Gaussian; based on the auxiliary prior information, generating a group of compact Sim3 motion bases, and obtaining each three-dimensional Gaussian transformation function in the dynamic scene image by using the weighted combination of the Sim3 motion bases, the parameters of the three-dimensional Gaussian transformation function including the transformation of the position and the shape along with the time; and performing three-dimensional Gaussian transformation to each frame of image of the dynamic scene according to the three-dimensional Gaussian transformation function, and performing image rendering on each frame to obtain a rendered image. According to the method, monocular depth and dense displacement field prior are combined, and the problem of monocular scale drift is solved by utilizing depth relative sequence stability. According to the method, real-time rendering is ensured, and meanwhile, the dynamic scene reconstruction precision is remarkably improved.

The invention discloses a unified monocular three-dimensional target detection method based on a state space model backbone network, and belongs to the technical field of computer vision. Comprising the following steps: acquiring an image data set, and labeling; a unified monocular three-dimensional target detection model is constructed and comprises a backbone network, a neck network, a multi-head feature processing module, a BEV feature generation module, a BEV encoder module, a BEV decoder and a detection module which are connected in sequence; training a unified monocular three-dimensional target detection model by using the image data set; and realizing unified monocular three-dimensional target detection by using the trained unified monocular three-dimensional target detection model. According to the method, the visual backbone network of the state space model is applied to the unified monocular three-dimensional target detection task, and the calculation complexity can be reduced to the linear level while the long-distance dependency relationship can be effectively modeled. Therefore, when sequence data and image data are processed, the potential of high performance and high efficiency is achieved.

The invention discloses an illumination adaptive image preprocessing method and system for monocular Gaussian sputtering SLAM, and belongs to the technical field of computer vision and robots. Aiming at the problem that the front end of the existing SLAM system depends on a luminosity consistency hypothesis, and positioning drift and mapping artifacts are easy to generate under intense illumination fluctuation, the method decouples a processing flow into a global dimension and a local dimension. In a CIELAB color space, smooth correction of global exposure is realized through a continuous interpolation algorithm, a multi-factor joint constraint mechanism is introduced to adaptively calculate a dynamic threshold value of local contrast enhancement, and intermediate-frequency edge information is directionally improved in combination with a Gaussian difference sharpening technology. According to the method, the robustness of the system is enhanced in an illumination abrupt change and extremely dark environment, the SLAM positioning precision is remarkably improved, the image rendering capability is improved when 3D Gaussian sputtering is used for scene reconstruction, and suspended noisy points and geometric cavities in a map are effectively eliminated.

The utility model relates to the technical field of single-tube telescopes, and discloses a portable single-tube telescope, the single-tube telescope adopts a supporting integrated structure arrangement, a protective cover structure is installed and connected by using a telescopic support arm structure, the structure is simple, the layout is compact, the structural volume is small after the single-tube telescope is stored, and the single-tube telescope is convenient to carry. The cover structure is composed of a connecting frame seat and a protective turning cover, the protective turning cover is installed on the front portion of the connecting frame seat in an overturning mode through cooperation of a rotating shaft support and a limiting clamping block limiting fixed shaft, when the device is held and used, the supporting arm does not need to be unfolded, the objective lens can be opened only through overturning, and holding use and fixed-point supporting work of the device are completed; the protective flip cover serves as a bottom support structure, fixed-point placement supporting of the telescope body can be achieved through matched supporting of the flip supporting arm and the telescopic supporting arm, the telescope body can meet different work use requirements of movable holding and fixed-point placement, and the use flexibility and the work scene applicability of the device are greatly improved.

1. The name of the design product: thermal imaging monocular. 2. The use of the design product: infrared imaging telescope. 3. The design points of the design product: in shape. 4. The picture or photo that best indicates the design points: perspective view.

1. The name of the design product: monitor (Q1 monocular MINI ball machine). 2. The use of the design product: the design product is used for monitoring. 3. The design points of the design product: in shape. 4. The picture or photo that best indicates the design points: perspective view 1.

Aspects of the invention disclose a variable-magnification monocular, comprising an objective lens group, a prism group and an eyepiece group arranged in sequence along an optical axis direction, wherein the eyepiece group comprises an eyepiece compensation group, an eyepiece variable-magnification group and an eyepiece fixing group arranged in sequence; the focal lengths of all lens groups satisfy the following conditional formula: −1.4<F31 / F32<−0.9 and 2<F33 / F32<1.7. The eyepiece magnification group and the eyepiece compensation group may move closer to or farther from each other along the optical axis to achieve continuous change of magnification between 60-20. The monocular may achieve continuous adjustment of the magnification of the monocular by moving the eyepiece zoom group and the eyepiece compensation group closer to or farther from each other along the optical axis, which can meet the needs of different observation objects.

The invention discloses a medical bed motion tracking system and method based on trinocular vision. The system adopts a non-equidistant compact three-camera layout, the left and middle base line is about 30cm, the left and right base lines are about 40cm, and all-angle dead-angle-free medical bed pose capture is realized by matching with medical spherical reflective mark points. Aiming at the shielding problem in a clinical environment, a dual anti-shielding strategy is adopted, trinocular redundancy characteristics are physically utilized, and binocular tracking is automatically switched when a monocular is shielded; a marker ball template is introduced in algorithm to match and repair damaged features. In a stereo matching link, an epipolar line and antipolar line dual verification mechanism is provided for a one-to-many mismatching problem, mismatching is eliminated through bidirectional projection consistency, uniqueness and accuracy of a matched triple are ensured, and high-precision three-dimensional reconstruction and pose tracking are further realized.

The utility model relates to the technical field of single-tube telescopes, and discloses a single-tube telescope with a magnetic attraction structure, one end of a single-tube telescope body is provided with an eyepiece, and the other end of the single-tube telescope body is provided with an objective lens; a storage assembly is arranged on one side of the single-tube telescope body and is used for accommodating accessories; a connecting assembly is arranged on one side of the storage assembly; the storage assembly comprises a box body, a box cover is hinged to one side of the top of the box body, automatic closing is achieved through attraction between the box cover and a first magnet on the top of the box body through an iron sheet, a user does not need to manually conduct complex locking operation, the first magnet and the iron sheet can be automatically attracted only by turning the box cover downwards to make contact with the box body, and the box cover can be automatically closed. Space partitioning is achieved in the box body through plug-in connection of the limiting grooves and the partition plates, and a user can flexibly adjust the positions of the partition plates according to actual requirements, so that the size and the shape of the containing cavities are changed, appropriate storage spaces are provided for different accessories, and diversified storage requirements are met.

1. Name of the product in this design: Camera (Monocular and Binocular Camera). 2. Intended use of this design: for use in vehicle-mounted or aircraft-mounted cameras. 3. The key design feature of this product is its shape. 4. The image or photograph that best illustrates the key design points: Design 1 3D view. 5. Design 1 is designated as the basic design.

This invention relates to a method for calculating the impact response spectrum based on hybrid monocular and binocular high-speed video measurement, comprising the following steps: camera calibration of the high-speed camera to obtain camera parameters and scale factor; stereo matching of the initial static frames of the binocular images obtained by the high-speed camera under static conditions; identification of occluded targets in the left and right image sequences using an affine transformation-based occlusion target detection algorithm; image sequence matching based on the occlusion target detection results, performing binocular stereo-sequence dual matching for unoccluded speckles, and monocular sequence matching for speckles occluded only in one camera view; and, based on the image sequence matching results, sequentially performing coordinate calculation, coordinate system correction, information fusion, displacement and acceleration calculation, and impact response spectrum calculation to obtain the impact response spectrum. Compared with existing technologies, this invention has advantages such as accurate impact response spectrum calculation results under occlusion conditions.

1. Name of the product in this design: Handheld Monocular Infrared Thermal Imager. 2. Application of this design: This design is a handheld monocular infrared thermal imaging device, mainly used for observation in scenarios such as safety monitoring, search and rescue, and outdoor exploration. 3. The key point of the design of this product lies in its shape. 4. The picture or photo that best illustrates the design points: Stereoscopic drawing 1.

The invention belongs to the technical field of optical system precision detection, and particularly relates to a binocular-assisted single-objective telecentric structured light system calibration and measurement method and system. The method comprises the following steps: obtaining a phase difference value and a height difference value of a reference plane based on a measurement system, fitting a mapping relation between the phase difference value and the height difference value, and realizing calibration from phases to heights of two monocular targets; and converting the phase value of the to-be-measured object into a height value and a point cloud coordinate based on the fitted mapping relation between the phase difference value and the height difference value, and realizing three-dimensional reconstruction and measurement of the to-be-measured object through point cloud splicing. According to the method, the symbol ambiguity problem of a monocular camera lens calibration method is solved, meanwhile, the precision and efficiency problems brought by a traditional projector calibration method are avoided, the phase matching link of an object is avoided, and therefore the problems of matching errors and information loss caused by phase matching under the shielding condition are effectively solved.

The present application discloses a monocular handheld observation device. The disclosed monocular handheld observation device comprises a cylindrical housing, a lens, an image sensor and an adjustment mechanism, wherein the lens is arranged at a second end of the cylindrical housing, the image sensor is arranged in the cylindrical housing and is located between a first end of the cylindrical housing and the second end of the cylindrical housing, and the lens and / or the image sensor is movably arranged on the cylindrical housing. The adjustment mechanism comprises a transmission assembly and an operating member, wherein the operating member is movably arranged at an observation end, at least part of the operating member is exposed outside the cylindrical housing, the operating member is connected to the lens and / or the image sensor by means of the transmission assembly, and by means of the transmission assembly, drives the lens and / or the image sensor to move, so as to adjust the focal length between the lens and the image sensor. The cylindrical housing is provided with a holding portion, and the holding portion is located between the lens and the operating member. The solution above can solve the problems in the related art of low flexibility and poor portability of monocular handheld observation devices during use.

This invention discloses an optically optimized, non-rotating curved mirror single-lens AR-HUD optical path system. The system includes an image source, a primary mirror, a secondary mirror, and a windshield. Light rays from the image source pass through the primary mirror, secondary mirror, and windshield to create a virtual image in front of the windshield, which is observed at the virtual image viewing position represented by the eyepiece. The secondary mirror is fixedly mounted on a support bracket of the AR-HUD optical path system. The system utilizes a free-form surface design based on multiple ray tracing simulations and optimizes the surface coating using a multi-wavelength optimization method. This invention also discloses a large-size single-lens box ray path optimization method with broad application value.

The invention provides an adaptive lane line detection method and device based on a foresight monocular camera, and the method comprises the steps: carrying out the illumination adaptive preprocessing of an RGB image collected by the foresight monocular camera, and obtaining a preprocessed image with balanced illumination, so as to solve a problem that the extraction of a lane line is not stable because of the uneven illumination; inputting the preprocessed image into a lightweight hierarchical feature extraction module to obtain a multi-scale fusion feature map, and considering the real-time requirements of long-distance narrow and thin lane line detection and a low-computing-power chip; then, based on the fused feature map, lane line key point coordinates are output through a dynamic key point prediction module, and the detection precision and the frame rate are balanced; and finally, performing curve fitting on the coordinates of the key points, and outputting lane line parameters. The method can improve the backlight shadow scene detection effect and the long-distance lane line recall rate, is suitable for sharp turn scenes, and meets the requirements of an ADAS system.

The application discloses a kind of equipment and method for functional valve seal assembly quality visual inspection.It includes monocular color industrial camera, plane diffuse light source, camera light source fixing plate, two kinds of workpiece positioning devices and other accessories.Monocular color industrial camera can provide good illumination environment for monocular color industrial camera image acquisition;In addition, by being designed with two kinds of workpiece positioning devices in the application, the same equipment can simultaneously meet the inspection needs of large quantities of functional valve seals and adapt to new models of functional valve seals, realize the automatic acquisition, analysis and determination of quality information of seal rubber ring and protective ring as detection feature points, solve the problem of misloading and missing loading, greatly reduce the limitation that traditional manual inspection can only check one sleeve shaft rubber ring at a time, shorten the inspection time and retain product quality certificate function, effectively improve the detection quality and efficiency of sleeve shaft rubber ring assembly.

The utility model relates to the technical field of telescopes, in particular to a telescope hinge structure and a telescop.The telescope hinge structure comprises two lens cone bodies, a connecting plate is inserted into the inner sides of the lens cone bodies, a middle shaft rod is slidably arranged in an inner cavity of the connecting plate in a sleeved mode, a containing groove is formed in the middle of the top of the connecting plate, and the middle of the middle of the top of the connecting plate is provided with a telescopic rod. A sponge pad is placed in an inner cavity of the storage groove, an oil discharge hole is formed in the bottom of the inner cavity of the storage groove, a mounting insertion plate is fixedly connected to the outer side of the connecting plate, and a positioning mechanism is arranged on the inner side of the lens cone body and comprises a mounting groove. The push plates on the two sides are pushed inwards, the positioning insertion plates are moved out of the mounting insertion grooves, the lens cone body is pulled outwards, so that the hinge can be independently maintained, the maintenance is more convenient, the lens cone body can be independently replaced, the replacement cost is saved, meanwhile, the lens cone body after being disassembled can be used as a single-tube telescope, and the practicability is improved.

This disclosure provides embodiments of a method, apparatus, device, and computer-readable medium for generating gesture depth information. One specific implementation of the method includes: controlling an associated monocular RGB camera to acquire a target gesture image; performing gesture keypoint detection on the target gesture image to obtain a set of keypoint coordinate information, wherein the keypoint coordinate information set corresponds to the gesture region included in the target gesture image; generating a set of bone length information based on the set of keypoint coordinate information; inputting the bone length information set into a pre-trained hand shape parameter generation model to obtain hand shape parameter information; and for each keypoint coordinate information set in the set of keypoint coordinate information, generating gesture depth information based on the keypoint coordinate information set, the hand shape parameter information, and the bone length information set. This implementation can reduce the power consumption of the camera device and simplify user operation.

1. The name of the design product: camera (monocular). 2. The use of the design product: for monitoring and video recording. 3. The design points of the design product: in shape. 4. The picture or photo that best indicates the design points: perspective view.

The invention discloses a monocular geometric estimation model reasoning acceleration method, which comprises the steps of performing feature extraction on an input image to generate an initial feature map, calculating a regional complexity score of the feature map, dynamically generating a sampling interval for attention calculation according to the complexity score, and calculating a monocular geometric estimation model. And sparse sampling is carried out on the key sum value based on the sampling interval to reduce the attention calculation amount, and feature coding is completed through the transformed attention layer to output a geometric estimation result. The method has the advantages that the calculation amount is reduced through dynamic sampling, the reasoning process is accelerated, and meanwhile the estimation accuracy is kept. According to the method, reasoning acceleration of monocular geometric estimation is efficiently realized.

This invention discloses a method and apparatus for fusion of monocular and binocular structured light 3D data. The method includes: S1, individually calibrating a binocular structured light 3D reconstruction system, a left monocular structured light 3D reconstruction system, and a right monocular structured light 3D reconstruction system; S2, jointly calibrating the individually calibrated binocular, left monocular, and right monocular structured light 3D reconstruction systems; S3, dividing the view area into region B and region M based on the views captured by the left and right cameras; and performing 3D reconstruction of the divided regions based on the binocular, left monocular, and right monocular structured light 3D reconstruction systems to obtain 3D point cloud data; S4, performing rigid body transformation on the 3D point cloud data and merging them to obtain fused monocular and binocular 3D point cloud data. This invention solves the problems of missing and hole-like points in the binocular structured light 3D reconstruction data caused by the difference in the field of view of the left and right cameras.

The invention relates to the technical field of image processing and three-dimensional computer vision, solves the technical problem of low precision caused by model noise and recognition algorithm limitation when coordinates of a mark point are directly extracted from a three-dimensional scanning model in the prior art, and particularly relates to a monocular 6DOF measurement-oriented three-dimensional mark point automatic detection and calibration method. Comprising the steps that a three-dimensional grid model of a measured target and a texture mapping file corresponding to the three-dimensional grid model are obtained, and a plurality of mark points with unique identity codes ID are arranged on the measured target; loading the three-dimensional grid model into a processing system, rendering the three-dimensional grid model through a virtual camera to obtain a projection image, and performing mark point detection to obtain an initial 3D coordinate; according to the method, high-precision geometric prior information provided by a three-dimensional scanner is fully utilized, optimization calibration is carried out through multi-view constraint of a real image, and the precision of the finally obtained mark point coordinates is remarkably higher than that of any single method.

1. Name of the design product: monocular holder. 2. Use of the design product: used for security monitoring camera. 3. Design points of the design product: in shape. 4. Pictures or photos that best show the design points: perspective view.

The application provides a monocular endoscope stereo imaging system and a three-dimensional measurement method, and relates to the technical field of endoscope imaging. The monocular endoscope stereo imaging system comprises a monocular endoscope body and an optical prism body, the monocular endoscope body is internally provided with an imaging unit and a lens unit, first incident light and second incident light correspondingly form first image information and second image information on the imaging unit after being reflected by the optical prism body, the first image information and the second image information have horizontal parallax, and the first image information and the second image information after stereo correction have no vertical parallax. The monocular endoscope stereo imaging system provided by the application can realize stereo imaging based on a single camera structure, overcome the defects of the prior art, reduce the hardware size, reduce the development cost, and improve the image quality.

A monocular blurred video reconstruction method based on 4D Gaussian splatting belongs to the fields of image restoration and three-dimensional reconstruction. This method addresses the problems of existing monocular blurred video reconstruction methods, such as low video visual quality and poor real-time rendering performance, when processing object motion blur. The present invention constructs a 4DGS model, simulates the physical formation process of camera motion blur and object motion blur, and synthesizes blurred images. This is used to calculate the loss with real blurred images to train the constructed 4DGS model, and then uses the trained 4DGS model to achieve video reconstruction. The present invention is primarily used to achieve 4D scene reconstruction.

The application discloses a light adaptive image preprocessing method and system for monocular Gaussian sputtering SLAM, and belongs to the field of computer vision and robot technology. In view of the problem that the front end of the existing SLAM system relies on the assumption of photometric consistency, positioning drift and mapping artifacts are prone to occur under severe light fluctuation, the processing flow is decoupled into two dimensions of global and local. In the CIELAB color space, the global exposure is corrected by a continuous interpolation algorithm, a multi-factor joint constraint mechanism is introduced to adaptively calculate the dynamic threshold of local contrast enhancement, and the Gaussian difference sharpening technology is combined to directionally improve the medium frequency edge information. The method enhances the robustness of the system under light mutation and extremely dark environment, significantly improves the SLAM positioning accuracy, improves the image rendering capability when using 3D Gaussian sputtering for scene reconstruction, and effectively eliminates the suspended noise points and geometric holes in the map.

This invention discloses a panoramic 3D measurement method and system for highly reflective aerospace components based on reference objects and monocular / binocular fusion, belonging to the field of 3D measurement and machine vision technology. The method includes the following steps: constructing a multi-view measurement system composed of a projector and a camera, and rigidly fixing the aerospace component to a stereo reference object with a globally unique ID; performing multi-view scanning by changing the relative pose of the measurement system and the measured assembly, obtaining robust phase information at each viewpoint using a multi-step phase-shifting method with pixel-by-pixel saturation frame culling, and reconstructing local point clouds using a pixel-level monocular / binocular point cloud fusion algorithm; simultaneously identifying and reconstructing the 3D coordinates of feature points on the reference object, and using the Kabsch algorithm to achieve high-precision alignment between the local point cloud and the global coordinate system; further repairing point cloud noise using an adaptive optimization algorithm based on normal vector noise evaluation, and finally fusing all aligned point clouds to obtain a complete high-fidelity 3D model of the aerospace component. This invention effectively solves the problems of missing data on highly reflective surfaces and difficulty in stitching together complex geometric features, achieving fully automatic high-precision 3D measurement without spraying or manual marking, significantly improving measurement efficiency and automation level.