63 results about "3d camera" patented technology

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

The invention discloses a visual localization and attitude estimation method and device based on prior search in a rocket recovery section and a medium, and belongs to the technical field of spacecraft guidance, navigation and control, and the method comprises the steps: carrying out the systematic sampling of a rocket recovery tail end operation space containing a three-dimensional position and a three-dimensional camera attitude in advance; constructing a matching data set in which the rocket recovery state is matched with the image; learning mapping from an image to a low-dimensional feature vector by using a deep learning model, and enabling the structure of a feature space to be consistent with the structure of a physical state space through a designed loss function; and in the rocket recovery stage, pre-stored samples closest to real-time image features are retrieved, and state labels of the pre-stored samples are fused, so that pose estimation is realized. The method can adapt to complex three-dimensional attitude changes, is high in robustness, and can meet the real-time requirement.

A hydraulic support roof support assembly device, system and method, comprising a transfer AGV, a workpiece grabbing robot, a feeding and discharging 3D camera, a conveying table, a multi-axis top mounting mechanism, a positioning 3D camera and a PLC controller; the transfer AGV serves as a moving main body; the workpiece grabbing robot is used for cooperating with the feeding and discharging 3D camera to grab and place a group of parts to be assembled into a roof support or a cover beam sleeve on a conveying table according to a set combination; the conveying table is used for conveying the group of parts to the multi-axis top mounting mechanism; the multi-axis top mounting mechanism is used for cooperating with the positioning 3D camera to realize positioning and angle adjustment, and push the group of parts into the corresponding roof support or cover beam sleeve. The present application does not need manual carrying or hoisting in the whole roof support or cover beam sleeve part assembly process, saves time and labor in assembly, has high automation degree, can greatly improve the assembly efficiency and precision, and meets the needs of the flexible assembly line of the hydraulic support.

A collimator control apparatus and method using artificial intelligence-based image processing and image analysis is provided. An exemplary apparatus includes a 3D camera configured to acquire a depth image of an imaging target; an RGB camera configured to acquire an RGB image of the imaging target; a collimator configured to adjust an irradiation field; a memory storing at least one process associated with controlling the collimator based on the depth image and the RGB image; and a processor configured to execute the process. The processor inputs the depth image and the RGB image into an artificial intelligence model to output a depth map, recognizes an imaging region and an imaging angle using the depth map and the RGB image, extracts a contour of the recognized imaging region to define a measurement region, and sets the irradiation field of the collimator using the measurement region.

Computer system (110), comprising: a communication interface (113) configured to communicate with at least one camera, comprising a first camera (270) with a first camera field of view (272); a control circuit (111) configured, when a stack (250, 750) with multiple objects is located in the first camera field of view (272), to: receive camera data generated by the at least one camera, wherein the camera data describes a stack structure for the stack (250, 750), the stack structure being formed from at least one object structure for a first object of the multiple objects;Identify, based on camera data generated by the at least one camera, a target feature (251B, 251C, 751B) of the object structure or a target feature (251B, 251C, 751B) located on the object structure, wherein the target feature (251B, 251C, 751B) is at least one of the following: a corner (251B) of the object structure, an edge (251C) of the object structure, a visual feature (751B) located on a surface (251A, 751A) of the object structure, or an outline of the surface (251A, 751A) of the object structure; Determine a two-dimensional, 2D, region (520, 620, 720) that is coplanar with the target feature (251B, 251C, 751B) and whose boundary is defined by the target feature (251B, 251C, 751B); Determining a three-dimensional, 3D, region (530, 630, 730) defined by connecting a location of the first camera (270) and the boundary of the 2D region (520, 620, 720), wherein the 3D region (530, 630, 730) is part of the first camera's field of view (272);Determine, based on the camera data and the 3D region (530, 630, 730), a size of an occlusion region (570, 670, 770), where the occlusion region (570, 670, 770) is a region of the stack structure located between the target feature (251B, 251C, 751B) and the at least one camera (270) and within the 3D region (530, 630, 730); Determine a value of an object detection confidence parameter based on the size of the occlusion region (570, 670, 770); and performing an operation to control robot interaction with the stack structure, wherein the operation is performed based on the value of the object recognition confidence parameter; wherein the first camera (270) with which the communication interface (113) is configured to communicate is a 3D camera configured to generate, as part of the camera data, several 3D data points that specify corresponding depth values ​​for locations on one or more faces of the stack structure.

A 3D planning table for collaborative layout planning, characterized by a flexible, rollable display (e.g., OLED display or higher) on which additively printed 2.5D miniatures of planning objects (e.g., machines, workstations, etc.) are placed. The current layout state is continuously captured using a 2D or 3D camera and transmitted via an interface to a CAD program, which translates the 2.5D representation into a 3D representation of the layout and displays it on a monitor and via VR glasses. Saved intermediate states can be transferred back to the OLED display via an interface between the CAD software and the planning process. All 2.5D miniatures are equipped with a camera-readable code, which allows further information to be retrieved from a central database using a smartphone camera.

A system for setting up a vision system to analyze and determine dimensions of cuboidal objects passing on a moving surface through a scene imaged by a 3D camera assembly with an image sensor, the system comprising: a user interface process that drives a user interface with a plurality of interface screens for setup and runtime operation of the system, at least one of the interface screens providing a setup display for determining optimal distance between the image sensor and the conveyor surface based upon parameters of the camera assembly, speed of the moving surface and width of the moving surface and a range of minimum and maximum size measurements for the object, wherein the at least one interface screen displays the optimal distance between the image sensor and the conveyor surface and displays an actual distance between the image sensor and the conveyor surface.

This invention provides a 3D gesture recognition method, system, electronic device, storage medium, and program product. The method includes: in a near-field interaction mode, acquiring a first 3D image containing a hand captured by a 3D camera; obtaining first 3D position information of the hand based on the first 3D image; obtaining a first 2D image based on the first 3D image; obtaining hand joint information based on the first 2D image; and determining 3D position information of the hand joints based on the first 3D position information and the hand joint information to determine gesture information; in a far-field interaction mode, acquiring a second 3D image containing a hand captured by a 3D camera; acquiring a second 2D image containing a hand captured by a 2D camera; obtaining second 3D position information of the hand based on the second 3D image; obtaining hand joint information based on the second 2D image; and performing coordinate fusion on the second 3D position information and the hand joint information to determine the 3D position information of the hand joints to determine gesture information.

Embodiments of the present application provide a large gantry type identification method and device based on 3D vision, which comprises: acquiring a point cloud set of the upper surface features of a workpiece moving at a constant speed through a 3D camera; filtering out background point clouds in the point cloud set to obtain preliminary extracted point clouds; through statistical filtering, removing random noise points distributed globally from the preliminary extracted point clouds, and then through clustering segmentation, eliminating invalid point cloud clusters, and obtaining pure point clouds; calculating the minimum bounding box of the pure point clouds, and determining the main direction of the workpiece and establishing a workpiece coordinate system; in the constructed workpiece coordinate system, performing feature calculation on the workpiece; comparing the results of the feature calculation with a pre-stored all-gantry-type-feature database, and outputting the gantry type according to the comparison result. Through complete point cloud reconstruction, multi-feature fusion identification, size and structure double verification, and man-machine collaborative error correction mechanism, high-precision, high-flexibility and high-robustness automatic identification of large gantry types is realized.

The application discloses a kind of hexahedron measured piece detection method, device, system, equipment and storage medium.Method includes: obtaining the image corresponding to six faces of hexahedron measured piece, six faces include front, back and four sides;Front corresponding image and back corresponding image both include the image obtained by photographing under backlight respectively and the image obtained by photographing under annular light source;3D image of hexahedron measured piece is obtained by 3D camera;According to the detection result whether six face corresponding image of hexahedron measured piece exists defect, according to 3D image, determine the detection result whether the height of hexahedron measured piece is qualified.The technical scheme can automatically complete the defect detection of hexahedron measured piece, improve detection efficiency and accuracy.

This teaching relates to estimating 3D camera pose based on 2D features detected from 2D images. A virtual 3D camera pose is generated relative to a 3D model of a target organ and associated anatomical structures. Virtual 2D images are created by projecting the 3D model from a viewpoint determined based on the virtual 3D camera pose. Each virtual 2D image includes the 2D structure of the target organ and / or some 3D anatomical structures visible from the corresponding viewpoint. A 2D feature / camera pose mapping model is then obtained based on the 2D features extracted from the virtual 2D images and the corresponding virtual 3D camera pose, where the 2D features include 2D ridge lines projected from 3D ridges on the target organ represented in the 3D model.

Embodiments of the present application provide a 3D camera and a debugging method and device, a storage medium and a program product. An industrial computer acquires a point cloud diagram collected by a laser line scanning structured light camera (3D camera) based on a current laser projection angle and a corresponding depth diagram. In response to a laser debugging operation, the industrial computer determines a target laser projection angle according to a to-be-debugged position marked in the depth diagram and camera calibration parameters, sends control information including the target laser projection angle to the laser line scanning structured light camera to control the laser line scanning structured light camera to project a laser line based on the target laser projection angle, so that the laser line can cover the to-be-debugged position, and the point cloud imaging effect of an area where the to-be-debugged position is located can be accurately improved. The user does not need to manually and repeatedly adjust camera parameters, the efficiency and accuracy of 3D camera debugging can be improved, and thus the point cloud imaging effect is improved.

This application discloses a rapid deployment intelligent welding device for collaborative robots, belonging to the field of robotic welding technology. It includes: a guide rail, which is mounted on the workpiece to be welded. A sliding component is slidably connected to the guide rail, and the sliding component is connected to the collaborative robot. It also includes a welding component, which includes a welding torch disposed at the end of the collaborative robot's arm, electrically connected to the collaborative robot. Moving the sliding component allows the collaborative robot to move along the guide rail, thereby enabling welding operations on large workpieces. An intelligent 3D camera is also mounted on the collaborative robot's arm, used to capture images of the weld seam. When the collaborative robot moves to a position near the weld seam via the sliding component, the intelligent 3D camera can scan and extract the precise location of the weld seam and transmit this location information to the collaborative robot. When automatically welding the same weld seam at different locations, precise welding can be achieved without repeated teaching.

Ways to facilitate multimodal interaction and communication for users are provided. A personal assistant system for a user includes a portable device equipped with a three-dimensional (3D) camera for generating images of 3D space and a propulsion system for self-propulsion. The personal assistant system also comprises a speaker, a microphone, and a control system interfacing with these components. The control system features a navigation module for controlling the propulsion system, an audio recognition module for converting audio from the microphone, and an image recognition module for processing images from the 3D camera. A processor transforms these formats into a processor-output, while an artificial intelligence (AI) module analyzes the formats to identify actionable commands and convert them into processor-output. An output generating module then converts the processor-output into a user-friendly output. Use of the personal assistant system enhances user convenience, accessibility, and engagement in educational, professional, and personal pursuits.

The present application relates to a kind of multi-camera joint calibration method without public view, medium and equipment, by collecting the several two-dimensional images of several ball calibration targets and target tracking instrument conversion matrix;Several three-dimensional point clouds of calibration ball and ball center coordinates are collected;The ball center coordinates of robot end calibration ball and corresponding flange pose under tracking instrument coordinate system are collected;Two-dimensional ball center coordinates in each two-dimensional image are extracted, and three-dimensional ball center coordinates are converted to obtain the external parameter of each 2D camera;According to the coordinates of the ball center of each calibration ball, the external parameter of each 3D camera is solved;Space transformation closed loop equation is constructed;Linear solution of end tool parameter and robot base external parameter is solved, and linear solution is used as initial value to construct the nonlinear least square optimization problem with the goal of minimizing space transformation closed loop equation residual, and the optimal solution of end tool parameter and robot base external parameter is solved, and calibration is completed.Thereby the problem of high-precision collaborative work of multi-source sensor in complex intelligent manufacturing scene is improved.

The application discloses a multi-sensor-based AGV I-beam wheel loading and unloading pose calibration method and system, the method comprising: collecting three-dimensional point cloud data of a hanging bracket cantilever through a 3D camera; scanning annular profile data of an I-beam inner hole through a 2D laser sensor; preprocessing the three-dimensional point cloud data and the annular profile data and converting them to the same vehicle body coordinate system through a coordinate system; executing a double-target pose collaborative matching algorithm based on the converted data to determine the relative pose deviation of the hanging bracket cantilever and the I-beam; including: performing feature extraction and feature constraint matching on the converted data, and the constraints including a circle center alignment constraint and an axis parallel constraint, calculating the relative pose deviation of the hanging bracket cantilever and the I-beam based on the matching features; and performing dynamic deviation real-time calibration according to the relative pose deviation, adjusting the vehicle body position or the gear cutting angle. The application can improve the accurate and stable AGV I-beam wheel automatic loading and unloading, and improve the efficiency and quality of the entire industrial production.

1. Name of the product in this design: Line Laser 3D Camera. 2. Purpose of this design: For the three-dimensional measurement and defect detection of industrial parts. 3. The key design feature of this product is its shape. 4. The image or photograph that best illustrates the design's key points: a 3D model.

This invention provides a parking lot toll gate system with vehicle location recognition and self-adjusting toll collection, belonging to the field of parking lot management technology. It solves the technical problems of existing parking lots' inability to stably, efficiently, and accurately collect tolls. The system includes a cloud-based management platform layer, an edge gate processing layer, a field perception execution layer, a multi-dimensional perception acquisition layer, an edge computing gateway layer, a cloud-based intelligent processing layer, a dynamic billing decision layer, an anti-evasion verification layer, a gate control execution layer, and a user interaction layer. The field perception execution layer includes a management booth, a self-locating barcode scanner, and auxiliary equipment. Weighing detectors are installed on both sides of the management booth, along with entrance / exit barriers, a second-level barrier, and a 3D camera on both sides. The self-locating barcode scanner is positioned above the weighing detector at the parking lot exit. This invention enables accurate vehicle location recognition, adaptive toll collection and release adjustment, reliable edge-end anti-evasion measures, convenient contactless payment, and cloud-based intelligent optimization.

The present application relates to diesel generator set test equipment technical field, specifically for remote management cabinet integrated diesel generator test load air inlet alarm control device based on non variable displacement pump, including non variable displacement pump unit, air inlet baffle control unit, operation range monitoring unit, central processing unit and alarm unit, central processing unit receives starting signal, dynamic adjustment centrifugal fan speed and PID control air inlet baffle opening, ensure that the heat dissipation air volume meets the demand;Operation range monitoring unit acquires point cloud data through 3D camera, and the moving target in the warning area is identified by intrusion detection algorithm, and multi-stage alarm and safety interlocking forced shutdown are triggered;After test stops, calculate delay time to close air duct, and protect components cooling;The air pressure sensor monitors the air duct state in real time, and timely early warning blockage or leakage fault.The present application has high integration degree, precise heat dissipation, comprehensive safety protection, strong adaptability, effectively improves diesel generator test load operation stability and safety.

The present application relates to bone nail positioning and robot control method, system, electronic equipment and storage medium, relates to the field of information technology, including the following steps: obtaining 3D point cloud data containing bone nail through 3D camera; processing 3D point cloud data, identifying the three-dimensional space coordinates of each bone nail, and calculating the XYZ space coordinates of each group of bone nails with 6 to 9 bone nails as a group; the XYZ space coordinates of each group of bone nails are transmitted to the robot control device; the robot collaborative arm continuously moves according to the XYZ space coordinates, the industrial camera on the robot collaborative arm shoots 6 to 9 bone nails at a time, and all bone nail groups in the whole nail box are shot in turn through continuous movement; image recognition and OCR recognition are carried out on the image obtained by the industrial camera, and the digital information on each bone nail and the total number of bone nails are extracted. The present application solves the problems of low efficiency and easy error of traditional manual operation through automatic detection and operation process, and significantly improves the automation level, accuracy and efficiency of orthopedic surgery preparation.

The utility model belongs to 3D camera technical field discloses an integrated edge calculation function's intelligent 3D camera, including 3D camera module, embedded system and communication module, embedded system integrates CPU, GPU and NPU's processor, wherein CPU is responsible for scheduling, data acquisition and transmission, GPU is responsible for point cloud generation, NPU is responsible for edge calculation, through the collection of three -dimensional data inside the camera, point cloud generation and edge calculation, only the result transmission to external device after processing, significantly reduce data transmission and the processing burden of external device. The utility model compact structure, powerful, has provided efficient edge calculation solution for industrial field, can promote system efficiency and response speed, be applicable to the defect detection, size measurement, robot navigation etc. Variety of scene, have remarkable practical value and popularization prospect.

The application discloses an automatic image final inspection system for valve springs, comprising: an upper computer and a valve spring detection mechanism main body; the valve spring detection mechanism main body comprises: a feeding mechanism, a carrying mechanism, a 3D scanning mechanism, an end face detection mechanism, a sorting mechanism, a detection flow channel and a discharging mechanism which are installed on a workbench; the upper computer is used for analyzing received side images of valve springs shot by the 3D scanning mechanism and end face images of valve springs shot by the end face detection mechanism to generate defect types of the valve springs, and generating corresponding control instructions based on the defect types to control the sorting mechanism to sort the valve springs. The automatic image final inspection system for valve springs uses a 3D camera line scan and a surface array camera to perform multi-angle rapid shooting, combines a deep learning algorithm to realize defect detection of each aspect of the valve springs, and greatly improves detection efficiency and detection accuracy.

The application discloses a loading and unloading and de-stacking robot and a grabbing method, the grabbing method comprising the following steps: collecting stacking data by a 3D camera; segmenting a 2D image to obtain a segmentation result; fusing initial cloud data collected by a plurality of 3D cameras to obtain global point cloud data located in a same coordinate system; extracting the global point cloud data according to a mask, and denoising the extracted global point cloud data to obtain material point cloud data of each material; searching for a grabbing surface based on the material point cloud data, and calculating a grabbing pose of the grabbing surface; judging whether the material meets a multi-grabbing condition, merging the material point cloud data of the material meeting the multi-grabbing condition, and searching for the grabbing surface again based on the merged material point cloud data to update the grabbing pose of the grabbing surface; sorting the grabbing order of the material according to a grabbing strategy to obtain a grabbing order; and generating grabbing information of a mechanical hand grabbing a cargo box according to the grabbing order, the grabbing surface and the grabbing pose.

Embodiments of the present application provide a laser control method and a laser processing device, the laser processing device comprising a 3D camera and a laser. The method comprises: controlling the 3D camera to scan a workpiece to be processed according to a preset scanning path to obtain a global point cloud of the workpiece to be processed; processing the global point cloud and a preset workpiece processing pattern to obtain a three-dimensional processing track, determining a height of the laser relative to the workpiece to be processed or a focusing distance of the laser according to coordinate information and a mapping relationship of the three-dimensional processing track, and controlling the laser to process the workpiece to be processed according to the three-dimensional processing track and the height of the laser relative to the workpiece to be processed or the focusing distance of the laser. The present scheme provides a low-cost three-dimensional scanning scheme, which can efficiently complete three-dimensional scanning and reconstruction of the surface profile of the workpiece to be processed, greatly improves the efficiency and effect of laser processing, and is simple to operate and low in cost.

The present disclosure provides a tire mold online laser cleaning track planning method, which comprises: photographing the cleaning area of the mold by using a 3D camera to obtain point cloud data; extracting the contour line of the mold cleaning area by using the point cloud data; and completing the tire mold online laser cleaning and track planning by using the cleaning head to move according to the contour line. The present disclosure obtains point cloud data by using a 3D camera, obtains a contour line by using point cloud data, and completes the automation of tire mold online laser cleaning by using a cleaning head to move according to the contour line. The present disclosure also provides a tire mold online laser cleaning device.

The utility model requests protect an automatic feeding system, it includes feeding robot, feeding device, tray, tray sets up in feeding robot any one side, is used for temporary storage material, feeding device sets up in feeding robot any one side, is used for conveying material, the below of feeding robot is provided with and feeding robot electric connection's 3D camera, 3D camera is used for identifying the position of material and the position of feeding device and transmission to feeding robot, and feeding machine is used for grabbing material to feeding device. Feeding device and tray are placed around the periphery of the feeding robot, and the specific position is adjusted according to the on-site situation. The automatic feeding system of the present application can replace manual feeding of materials, thereby saving manpower, in addition, it can also avoid the influence of materials harmful to human health on human health.

The application discloses an intelligent robot system for assisting eye drop, aiming at solving the problems of inaccurate operation, poor compliance and inconvenient medicine management when patients with hand dysfunction, visual impairment and cognitive impairment autonomously drop eye drops. The system comprises a medical recliner, an auxiliary mechanical arm, a camera detection unit and a control unit. The camera detection unit comprises a 3D camera module and an eye movement tracking module, which can track the position of the eyeball in real time. The control unit dynamically adjusts the posture of the end of the mechanical arm accordingly, and realizes precise dripping. The system also integrates refrigerated storage, medicine identification, voice and display interaction and remote communication functions, and can automatically execute the whole process of taking medicine, dripping, homing and recording. The application realizes dynamic servo precise drug delivery, ensures the safety and compliance of drug use, and is suitable for long-term management of chronic eye diseases in family and medical scenes.