875 results about "3d camera" patented technology

The invention relates to a high-precision hand-eye calibration method for a robot system. The method comprises the following steps: planning a plurality of random robot poses; in the process of operating the pose of the robot, the coordinate value of a target ball point is measured through a laser tracker, an end effector tool coordinate system is constructed and converted into a robot base coordinate system, and an actual measurement rotation matrix and a displacement vector are obtained through calculation. In combination with the theoretical rotation matrix and the displacement vector, pose compensation is achieved through a rotation operator and a translation operator which are finely adjusted, and the target pose and displacement of the robot are adjusted; and after compensation, repeatedly measuring the actual measurement pose and calculating the residual error, and taking the residual error as a compensation value of next iteration until the residual error meets a preset precision threshold value or reaches the maximum number of iterations. And finally, recording the corresponding relation between the pose of the calibration plate in the 3D camera coordinate system and the theoretical pose of the robot to complete the robot hand-eye calibration process. According to the method, the calibration precision is improved, and the high-precision machining and assembling requirements are met.

The invention belongs to the technical field of mechanical arm automatic grabbing, and particularly relates to a 3D visual guidance mechanical arm grabbing method for irregular objects. Comprising the following steps that (1) a 3D visual guidance mechanical arm grabbing system is built, specifically, a 3D camera based on line structured light is adopted to guide a mechanical arm, the mechanical arm is a visual flexible six-axis mechanical arm, and after the mechanical arm completes accurate grabbing of all objects and carries out placement according to requirements, it is regarded that the grabbing task is successfully completed; and the camera in the 3D camera is calibrated, and the calibration of the camera is realized by calling a CalibrateCamera function in the OpenCV library. Through mutual cooperation of the 3D camera and the mechanical arm, intelligent grabbing of irregular objects such as industrial parts is achieved, more accurate grabbing operation is carried out according to the actual application situation, and the robot does not depend on training of three-dimensional point cloud samples. And the 3D camera can accurately obtain external information, and the grabbing requirement of the mechanical arm is met.

The invention relates to the technical field of mechanical automation, and discloses a novel part material intelligent production line which comprises a feeding truss provided with 3D scanning equipment and used for recognizing and positioning part materials; the beveling machine is provided with an X-axis moving assembly, a Y-axis moving assembly and a Z-axis lifting assembly and is used for adjusting the position of a cutter so as to meet the machining requirements of shapes, outlines and thicknesses of different parts; the chip removal system comprises a spiral chip removal machine and a plate chain chip removal machine and is used for collecting and conveying chips generated in the machining process; the feeding and discharging truss comprises an intelligent classification module and is used for recognizing the machined parts and placing the machined parts to designated positions according to categories. And the central control system is used for integrally scheduling and monitoring the production process. By means of the 3D camera system with the depth sensing and image analysis functions, high-precision part material recognition and positioning can be achieved at the initial stage of the production process. The accurate initial treatment lays a foundation for subsequent processing links, and ensures the accuracy and repeatability of the whole processing flow.

The invention discloses an element plugging test system based on visual perception and a control method and device.The system comprises a visual perception module, a grabbing device, a conveying device and an upper computer, and the visual perception module is used for conducting visual recognition on jacks of elements to be tested and comprises a 2D camera and a 3D camera; the 2D camera has a first field of view, the 3D camera has a second field of view, the first field of view comprises the second field of view, and the first field of view range is larger than the second field of view range; the grabbing device comprises a clamping end, and the clamping end is used for clamping the plug to plug and unplug the jack; the visual perception module is mounted at the clamping end so as to move along with the clamping end; the conveying device is used for conveying to-be-tested elements; the upper computer is in communication connection with the visual perception module, the grabbing device and the conveying device. The visual perception module can be compatible with large-deviation incoming materials, so that multi-degree-of-freedom position and posture adjustment is realized; therefore, the application range of the system is greatly expanded, and the system has excellent flexibility, compatibility and accuracy.

The invention discloses an access control face recognition system and method based on a three-dimensional technology, and relates to the technical field of access control face recognition, and the recognition system comprises a face data acquisition module which is used for capturing three-dimensional face data of a target person by using a 3D camera; the multi-modal fusion module is used for fusing the three-dimensional face data and the traditional two-dimensional image data; the modeling module is used for generating an accurate three-dimensional face model; the dynamic adjustment module is used for creating a real-time dynamic adjustment model; the matching module is used for extracting facial features from the three-dimensional face model and matching the facial features with the three-dimensional face model in the database; and the identity verification and authorization module is used for carrying out identity verification according to the matching result. According to the method, three-dimensional face data and traditional two-dimensional image data are fused, the recognition accuracy and reliability can be improved, a real-time dynamic adjustment function is introduced, recognition parameters can be automatically optimized according to different environment light rays and angles, and better use prospects are brought.

The invention provides a linear dimension measurement method and device for the total length of a road guardrail stand column, and relates to the field of linear dimension measurement, the device is deployed on a detection vehicle, and comprises a control module, a ground penetrating radar measurement module, a 3D camera measurement module, a sensor module, a lower computer and an upper computer; according to the invention, accurate data acquisition is realized through multi-module cooperative control, and the lower computer is combined with the AI image recognition technology to realize accurate acquisition of contour data; the lower computer transmits radar original data (reflected waves and diffracted waves) and 3D contour data (column cap space coordinates) to the upper computer, the upper computer achieves data fusion through a hierarchical velocity model (time-depth conversion) and a coordinate system mapping algorithm (3D-road coordinate system), and the accuracy of the total length measurement error is greatly improved compared with that of single module detection. The method solves the problems of low efficiency, large destructiveness and poor accuracy of traditional manual detection, and has the advantages of method innovation and engineering landing.

The present invention relates to the technical field of automobile curved screen detection, and in particular to a device and method for detecting the response time of an automobile curved screen, the method comprising: optimizing and affine transforming pixels on a two-dimensional plane image obtained by a 2D camera to obtain corresponding three-dimensional position information of the curved screen; determining the three-dimensional position information of a target icon on the automobile curved screen by a 3D camera, controlling a mechanical arm to accurately click the target icon on the automobile curved screen according to navigation coordinates, synchronously starting high-frame rate cameras set in different directions, obtaining continuous images and time information of changes in different areas of the curved screen when the mechanical arm operates on the automobile curved screen; inputting into a response time detection algorithm for analysis, and evaluating the response time performance of the automobile curved screen according to the output result of the response time detection algorithm. The present invention can realize accurate detection of the response time of an automobile curved screen.

Provided in the embodiments of the present application are a laser control method and a laser machining apparatus. The laser machining apparatus comprises a 3D camera and a laser. The method comprises: controlling the 3D camera to scan a workpiece to be machined according to a preset scanning path so as to obtain a global point cloud of said workpiece; processing the global point cloud and a preset pattern to be machined so as to obtain a three-dimensional machining track, and determining, on the basis of coordinate information and a mapping relation of the three-dimensional machining track, the height of the laser with respect to said workpiece or the focusing distance of the laser; and, according to the three-dimensional machining track and the height of the laser with respect to said workpiece or the focusing distance of the laser, controlling the laser to machine said workpiece. The present application provides a low-cost three-dimensional scanning solution, which can efficiently complete three-dimensional scanning and reconstruction of surface contours of workpieces to be machined, thus greatly improving laser machining efficiency and effect, and achieving simple operation and low cost.

The invention discloses a visual angle tracking grating stereo display system comprising a 3D camera grating display device used for obtaining viewer 3D information; the eyeball position tracking module is used for acquiring real-time spatial position coordinates of the eyes of the audience relative to the 3D camera; the pose calibration module is used for acquiring a spatial relationship between the 3D camera and the display; and the visual angle rendering and grating interlaced graph generation module is used for calculating the spatial position of the human eyes on the display screen and the mapping of each grating strip in a screen pixel coordinate system after acquiring the positions of the eyes and the relative poses of the display screen and the 3D camera, generating a grating stereoscopic display image suitable for the eyes, and sending the grating stereoscopic display image to a display to be output. According to the method and the device, the viewpoint self-adaptive three-dimensional display of the eyes of one or more audiences is realized, the calculation complexity is reduced, the three-dimensional vision consistency can be ensured, the picture which should be seen by the viewing angle can be accurately rendered, and the problem of blurring or 3D picture distortion is avoided.

The invention belongs to the technical field of machine vision fusion application, and particularly relates to a composite robot CNC feeding and discharging method and device. Comprising a composite robot and a CNC machine tool; the composite robot comprises a movable chassis, a mechanical arm, a feeding and discharging clamping jaw, a 3D camera and a 2D camera. The 3D camera and the 2D camera are mounted on a mounting platform at the tail end of the mechanical arm; the 2D camera and the 3D camera are added in the first deployment stage and the nth deployment stage to assist deployment, teaching of each CNC machine tool is not needed, the 3D camera is dismounted in the feeding and discharging stage, only the 2D camera is used in the feeding and discharging stage, and the cost is lower; and meanwhile, a 2D positioning identification plate or a 3D positioning identification plate is adopted, the feeding base does not need to be photographed, and the influence of cutting liquid, metal filings, discharging workpieces and the like on the camera algorithm precision is avoided.

Exemplary embodiments of the present disclosure are directed towards an indoor and outdoor navigation system for autonomous mobile robots and the methods employed therein. The system comprises a mobile robot equipped with sensors, including IMU, encoders, GPS, and a 3D camera, all managed by a processing device. During indoor navigation, the device utilizes an extended Kalman filter for precise odometry estimation, adapting to outdoor mode with an additional clearance base. User-defined waypoints guide the robot, supported by safety lanes generated through a lane generation algorithm to prevent off-roading in mapless environments. Visual markers facilitate odometry correction for enhanced accuracy, while map-based navigation employs 2D LIDAR sensors for environment mapping. Safety is ensured through obstacle detection and ground removal algorithms, with alerts transmitted via IoT protocols upon detecting obstacles beyond designated safety lanes, thereby enhancing navigation reliability and safety.

In accordance with one aspect of the inventive concepts, provided is a method and a system for horizontal infrastructure localization, comprising a robotic vehicle platform, such as an AMR, a mechanism for collecting sensor data, e.g., point cloud data, such as a LiDAR scanner or 3D camera, and a processor configured to process the sensor data to identify and determine a position and orientation of the horizontal infrastructure. The robotic vehicle processes the sensor data to identify a horizontal infrastructure, determine its pose, and then determine if an area of a horizontal surface of the horizontal infrastructure is clear for dropping a load, e.g., palletized load.

The invention is suitable for the field of image processing, and discloses a fast fusion method for point cloud data of multiple 3D cameras, terminal equipment and a storage medium. The multi-3D-camera point cloud data fast fusion method comprises the steps that calibration data of a plurality of 3D cameras are collected, the calibration data comprise an RGB image, a depth image and an internal reference matrix, and a calibration plate needs to appear in the view fields of the 3D cameras at the same time; generating a three-dimensional point cloud coordinate point set according to the calibration data, and calculating a transformation matrix between the reference camera coordinate system and the non-reference camera coordinate system according to the three-dimensional point cloud coordinate point set; and according to the transformation matrix, registering the three-dimensional point cloud coordinate point set to a preset reference camera coordinate system to obtain merged target point cloud data, the target point cloud data including RGB color values. According to the invention, multi-view complementation is realized, a physical blind area of a single camera can be covered, and the problem of point cloud missing caused by insufficient view field and shielding is solved.

The invention discloses an assemblage multi-robot collaborative welding system in ship sections, and relates to the technical field of industrial robot collaborative control, the assemblage multi-robot collaborative welding system comprises an industrial internet platform, the industrial internet platform is in communication connection with the following modules: a visual inspection modeling module, which is used for scanning the surface of a workpiece through a 3D camera, constructing a point cloud model, and establishing a point cloud model; matching the point cloud model with a design drawing, and identifying a deviation between an actual workpiece and the drawing; and the path planning and correcting module is used for planning a collision-free welding path in combination with the visual detection data and the motion range of the robot. Through high-precision visual inspection scanning and an SLAM matching algorithm, a workpiece point cloud model is constructed in real time and is accurately aligned with a design drawing, the welding seam position and the workpiece geometric deviation can be accurately recognized, a collision-free welding path is generated in combination with robot kinematics constraint and a safety boundary, it is ensured that the pose of the tail end of a welding gun is attached to the surface of the workpiece, and the welding quality is improved. Welding deviation is reduced, and process stability is ensured.

The invention discloses a visual positioning method for a high-environment self-adaptive mobile composite robot, which comprises an AGV trolley, a feature plate and a 2D camera, the AGV trolley is used for moving among stations and enabling a multi-axis mechanical arm located on the AGV trolley to pick and place a predetermined tool, the feature plate is arranged on the stations, the feature plate 1 is arranged on the tool, and the 2D camera is arranged on the multi-axis mechanical arm. The 2D camera is used for acquiring a 2D image; the visual positioning process comprises the steps that S1, a mechanical arm is controlled to shoot multiple sets of pictures at different angles (preferably 20 sets), the pose of the mechanical arm is obtained, then an internal parameter matrix and an external parameter matrix of a camera are calculated according to picture set data, and the parameter matrixes are used for 3D position calibration calculation. 3D poses (X, Y, Z, RX, RY and RZ) of material taking and placing points are obtained through a 2D camera; and the low-cost 2D camera is used for realizing partial functions of acquiring 3D poses by the 3D camera.

The invention relates to the technical field of industrial vehicle intelligence, in particular to an intelligent flexible forklift system based on multi-mode sensing and self-adaptive path planning. The intelligent forklift comprises an intelligent forklift body, a vehicle-mounted integrated control system, a multi-mode sensing and navigation module, a self-adaptive decision-making and path planning module, a cooperative communication and task management module and a cloud scheduling and management platform, and the intelligent forklift body is provided with a pallet fork mechanism with a 3D camera and a force sensor; the vehicle-mounted integrated control system serves as a core to achieve multi-module data interaction and instruction output, the multi-mode sensing and navigation module generates an environment situation map through multi-source data fusion, the self-adaptive decision-making and path planning module completes path planning, obstacle avoidance and pre-verification, the cooperative communication and task management module achieves multi-vehicle cooperation and task disassembly, and the multi-mode sensing and navigation module generates an environment situation map through multi-source data fusion. And the cloud scheduling and management platform realizes global monitoring and management. According to the method, the adaptability to complex scenes and the operation safety are improved, and accurate forking and flexible path planning are realized.

The invention relates to the technical field of welding, and discloses a teaching-free angle steel welding method which comprises the following steps that S1, a robot carries a 3D camera to execute self-adaptive high-angle shooting exploration, and workpiece point cloud data are obtained until recognition succeeds or exceeds a preset exploration range; s2, dividing the ROI into grids according to the view field of the camera by taking the distance of the point cloud detected for the first time as a reference; s3, performing expansion type multi-angle photographing from the initial grid to the periphery, stopping scanning when meeting an empty grid, and obtaining an effective area point cloud of the workpiece; s4, performing regional progressive registration on the point clouds of the adjacent shooting regions, and eliminating registration drift through closed-loop detection and pose image optimization; and S5, calculating the normal and curvature of the point cloud. According to the intelligent angle steel welding method, the robot carries the 3D camera to execute self-adaptive high-angle shooting exploration to obtain the point cloud data of the workpiece, and therefore the problems that due to the fact that manual teaching or off-line programming is mostly adopted in a traditional intelligent angle steel welding method, the welding efficiency is low, complex workpiece path planning consumes long time, and precision fluctuation is large are solved.

The invention relates to a method, a device and a system for detecting the damage of a container body. The method comprises the following steps: acquiring a 2D image of the container body acquired by a 2D camera; obtaining a plurality of groups of 3D point cloud data of the container body collected by a plurality of 3D cameras, wherein each group of 3D point cloud data corresponds to a local part of the container body; respectively registering each group of 3D point cloud data to a unified coordinate system to obtain complete 3D point cloud data of the container body in the coordinate system; and detecting the container body according to the 2D image and the complete 3D point cloud data. The invention further provides a corresponding container intelligent checking device and system. According to the method, the automation degree, efficiency and inspection quality of container inspection, the registration efficiency and accuracy of the 3D point cloud and the accuracy of 3D point cloud feature extraction can be improved.

A ship sub-assembly weld seam identification method, comprising: creating a point cloud weld seam identification training dataset by means of point cloud discretization, virtual sampling, weld-zone semi-automatic interactive annotation, and style transfer-based augmentation of ship sub-assembly CAD data; training a point cloud weld seam identification model, and exporting a weld seam identification model data file; using a 3D point cloud camera array as a sampling camera to collect workpiece 3D point cloud data of a sub-assembly workpiece to be welded at the production site; by means of the weld seam identification model data file, obtaining weld seam annotation position data and positioning a weld seam; and issuing a welding process plan and instruction to control a welding robotic arm and a control device therefor to complete sub-assembly workpiece welding. Also provided are a ship sub-assembly weld seam identification system, a computer device, and a computer-readable storage medium. The method overcomes the shortcomings in conventional 3D weld seam identification methods such as point cloud incompleteness and identification failures caused by steel types and the imaging angles of 3D cameras, thereby greatly improving the adaptability of automatic weld seam identification methods to complex workpieces.

The invention discloses a large-size complex surface type high-speed scanning measurement device based on a white light interference technology and a method thereof, and relates to the technical field of crossing of optical precision measurement and machine vision, and the device comprises an intelligent measurement module which integrates a large-view-field vision camera, a 3D camera and a white light interferometer, the large-view-field vision camera is used for acquiring a global two-dimensional image of a measured object and automatically identifying features to trigger a measurement program, and the 3D camera is used for acquiring three-dimensional point cloud data of the measured object; according to the invention, the intelligent measurement module integrates multiple devices, a large-view-field camera triggering program, point cloud acquisition by the 3D camera and high-precision measurement by the white light interferometer, so that multi-source data collaboration is realized. The dynamic scanning optimization module adjusts step lengths and grid priorities according to curvature regions, improves an algorithm to generate a path, improves scanning efficiency and pertinence, achieves combination of global and local measurement through cooperation of multiple devices, meets the multi-precision requirements of complex surface types, improves measurement efficiency, and guarantees the precision of key regions.

The invention relates to a railway track abrasion detection method and system, and aims to realize accurate and efficient detection of turnout sharp point rail abrasion. A detection device carrying a 3D camera, a speed sensor, a signal processing module and a data processing terminal is arranged in a turnout area to be detected, parameter configuration is completed, in the constant-speed moving process, the speed sensor generates a trigger signal according to the specified distance, the 3D camera is driven to synchronously collect point cloud data of the two-dimensional contour of the cross section of the sharp point rail, and a three-dimensional data stream is formed; after data are denoised, section registration splicing and three-dimensional model reconstruction are completed in combination with distance stamps, an actual measurement model and a standard contour model are superposed and compared, abrasion parameters are calculated, and the maximum abrasion point and position are recognized; and finally presenting the abrasion data in the forms of a color cloud picture, a table and the like, and automatically generating a standard detection report containing the contents of line information, defect types and the like. According to the method, automation and visualization of abrasion detection are achieved, the detection precision is high, the efficiency is high, and reliable data support is provided for track maintenance.

The utility model relates to the field of aircraft coating repairing technology research, in particular to an in-situ automatic repairing device for an aircraft local coating. Comprising a mobile operation platform, a 3D camera which is arranged on the mobile operation platform and is used for collecting 2D images and 3D point cloud data of the surface of an airplane, a repairing module which is arranged on the mobile operation platform and is used for coating cleaning or paint spraying, and an upper computer which is used for controlling the mobile operation platform, the 3D camera and the repairing module. Compared with the device in the prior art, the device can automatically inspect, identify and position the damaged area of the coating on the surface of the airplane, and in-situ automatic identification and repair of the damaged special-shaped bent coating on different parts of the surface of the airplane are realized.

The invention belongs to the technical field of unmanned aerial vehicle battery replacement, and provides an unmanned aerial vehicle automatic battery replacement method and system based on 3D visual guidance, and the method comprises the steps: constructing a target feature template of each region of a battery compartment and a feed battery based on obtained standard image data; performing multi-stage pyramid matching on the obtained to-be-matched point cloud data and the constructed target feature template, and reconstructing a point cloud data set and standard image data subjected to multi-stage pyramid matching; calculating the pose deviation value of the feed battery, and combining the pose deviation value of the feed battery to obtain the actual position of the feed battery; the target position and posture information are updated according to the feed battery posture deviation value, a mechanical arm is controlled to move to the target position according to the updated posture information, and the feed battery is grabbed and placed at the designated position; and controlling the mechanical arm to grab the fully-stored battery and then return to the specified image data acquisition position, controlling the 3D camera to carry out secondary scanning, repeating the steps, and putting the fully-stored battery into the battery compartment of the unmanned aerial vehicle to complete automatic battery replacement operation.

The embodiment of the invention relates to the technical field of automatic unstacking robots, and discloses an unstacking method and system, a controller and a computer readable storage medium, and the method comprises the steps: obtaining RGB image information and point cloud information of an ROI region through a 3D camera, recognizing a to-be-unstacked article in the RGB image information, and generating a grabbing point; the camera coordinates of the grabbing points, the length and width information of the box body of the top layer to be unstacked and the height information of the top surface of the box body are obtained through the point cloud information; and the grabbing step is executed, after the current grabbed box leaves the stack, the height difference of the original position of the current box is obtained through the 3D camera again, the detection height value of the current box is obtained, laser radar verification is selectively adopted according to different conditions, and finally efficient unstacking is achieved. By means of the technical scheme, the problem that in the unstacking process in the prior art, the detection efficiency and the operation precision are difficult to consider at the same time is solved.

The invention discloses a tubular steel body surface defect detection device and method, and relates to the technical field of detection equipment, the tubular steel body surface defect detection device comprises a detected steel pipe, a servo electric cylinder and a 3D camera, the 3D camera is arranged on the outer side of the detected steel pipe, the 3D camera is fixedly connected with the servo electric cylinder, and the servo electric cylinder is fixed on a device frame; the 3D cameras are respectively connected with the off-line computing system through wires; based on the equipment detection method, an on-line detection system and an off-line calculation system are included. According to the tubular steel body surface defect detection device and the detection method, non-contact measurement can be realized; online real-time detection is achieved, and the detection coverage range is widened; the detection precision is improved; the detection efficiency is improved.

The invention discloses an environment-friendly operation and maintenance robot vision system, and the system comprises an image collection module which comprises a 3D camera disposed on a mechanical arm of a robot, and is used for collecting a plane image and a depth image of a target object; the image processing module is used for preprocessing the plane image acquired by the image acquisition module and transmitting the plane image to the target detection module; the target detection module is provided with a rotating target detector obtained by training based on a YOLOv8OBB network model, and is used for detecting the image transmitted by the image processing module and outputting a target frame center point, a target frame width, a target frame height and a rotating angle; the data conversion module is used for obtaining a rotation translation matrix of the target relative to the 3D camera; and the result output module is used for converting the rotation translation matrix into a relative motion matrix of the mechanical arm and outputting the relative motion matrix to a mechanical arm control unit of the robot. Based on the scheme of the invention, an efficient and accurate water quality station robot vision system can be obtained.

The invention relates to the technical field of machine vision. The invention particularly relates to a workpiece grabbing collision interference detection method and system and a storage medium. The method comprises the following steps: acquiring a point cloud of a robot gripper at an initial position, and recording the point cloud as a first point cloud G of the robot gripper; obtaining a first point cloud S of a production scene through a 3D camera; a second point cloud # imgabs0 # of a production scene and a second point cloud # imgabs1 # of a robot gripper are obtained; obtaining a production scene point cloud depth map and a robot gripper point cloud depth map; obtaining an area in which the depth image Ims and the depth image Img both have projection depth values; obtaining the number of pixel points of which the depth values of the depth image Img are greater than the depth value of the depth image Ims; and in response to the fact that the number of the pixel points is larger than a preset collision threshold value, it is judged that the robot gripper will collide with other objects when grabbing the to-be-grabbed workpiece. By adopting the method provided by the invention, whether the gripper of the robot collides with the object in the production scene when moving to the grabbing point can be accurately judged.

The invention relates to a tire character recognition method, in particular to a tire character recognition method based on three-dimensional point cloud, and the method comprises the steps: calibrating an industrial 3D camera, managing and controlling the connection and communication of the camera through a tire point cloud data acquisition module, and visualizing and storing an acquisition result; performing combined filtering on the tire point cloud data by using voxel downsampling and a statistical filtering method, and then making a tire data set; putting the manufactured data set into a PointNet + + network added with a self-attention mechanism for training, and generating a training weight; and visualizing a tire identification result and storing a prediction result file. The invention relates to the technical field of three-dimensional recognition, geometric characteristics of the surface of a tire are more accurately captured by using point cloud, the influence of factors such as light and tread stains is reduced, hierarchical feature extraction of PointNet + + is combined with a Transform self-attention mechanism, detail features are more sufficiently extracted by using the relationship between different hierarchical features, and the recognition accuracy is improved. And the robustness of tire character recognition in a complex environment is improved.

The invention provides a cement conveyor volume measurement method based on deep learning point cloud segmentation. The method comprises the following steps: obtaining 3D point cloud data by using a laser line scanning 3D camera, and determining RGB values of pixels corresponding to the 3D point cloud data to obtain 3D point cloud with color features; inputting the 3D point cloud with the color features into an improved Point-Net + + neural network for 3D point cloud segmentation, and obtaining 3D point cloud data only containing the cement surface in the conveyor after segmentation; constructing a triangulation network based on the cement irregular surface in the segmented 3D point cloud data by adopting a triangulation method, and performing volume calculation according to the constructed triangulation network; wherein the improved Point-Net + + neural network is constructed on the basis of a framework of an original Point-Net + + point cloud segmentation network, and comprises a core point convolution, a context sensing module and a Focal Loss loss function. Therefore, the performance of a point cloud segmentation algorithm is improved, and even under complex conditions, the surface point cloud data of the cement material in the conveyor can still be accurately segmented for volume measurement.

A bearing defect detection system and method based on multi-combination lightsource and multi-view multi-channel YOLO network structure detection algorithm, in which the system includes a rotating platform, a multi-combination lightsource, a light source controller, a 2D camera and lens, an industrial control machine processing unit; The rotating mechanism is used for the rotation of the bearing at any 360 degrees, the light source adopts a set of adjustable coaxial or ring light sources, the camera is all 2D cameras, for pits, scratches, burrs and other three-dimensional defects, improved by multi-view multi-channel YOLO network structure detection algorithm, replacing the current expensive 3D camera detection methods, improve detection accuracy, reduce system costs.