1127 results about "Laser line" patented technology

The invention relates to the technical field of intelligent detection, in particular to a container surface damage detection method and device based on machine vision, and the device consists of a multi-modal data acquisition module, a dynamic compensation processing module, a multi-modal data fusion module and a damage classification and positioning module. The multi-modal data acquisition module generates a high-density three-dimensional point cloud through double-laser line scanning, and acquires a multispectral image, an infrared thermogram and a real-time motion state. The dynamic compensation processing module integrates an optical flow method and acceleration data to realize sub-pixel-level motion compensation, and combines adaptive exposure control to optimize the imaging quality under complex illumination. The multi-modal data fusion module strengthens defect feature expression through space-time alignment and a double-branch collaborative attention mechanism, and model parameters are reduced through lightweight network design. And the damage classification and positioning module adopts an improved deep network to realize defect classification and complete millimeter-level three-dimensional positioning. The automatic detection requirement of a port is met, and the problems of poor container detection precision, poor adaptability and the like are solved.

The invention belongs to the technical field of three-dimensional measurement, and discloses a 3D laser line scanning system based on a point cloud positioning algorithm and a control method. The method comprises the following steps: acquiring environment initial point cloud data and scanning target information; carrying out feature extraction, constructing an environment feature description model, carrying out three-dimensional space partitioning processing, carrying out regional reflectivity feature analysis, and constructing a space reflection feature model; adaptive optimization configuration is carried out on the laser scanning parameters, and a scanning strategy is generated; scanning path dynamic planning is carried out on the scanning target information, scanning density distribution calculation is carried out, and a scanning path is generated; performing laser line scanning control by using the scanning path and the scanning strategy, and performing scanning data quality real-time monitoring to obtain scanning quality evaluation data; scanning parameter iterative optimization is carried out on the scanning quality evaluation data, point cloud data real-time fusion reconstruction is carried out, a high-precision three-dimensional model is generated, the modeling precision is improved, and meanwhile the system resource utilization efficiency is optimized.

The invention discloses a welding gun pose online planning method for multi-layer and multi-channel external welding of a pipeline and related equipment. The method belongs to the technical field of intelligent welding and comprises the steps that a laser line image and welding gun posture data of a pipeline welding groove are obtained; extracting feature pixel points from the laser line image; converting the three-dimensional point cloud into a three-dimensional point cloud through a detection mathematical model, fitting to obtain a three-dimensional welding groove graph, and calculating a welding groove size parameter, a welding gun pose parameter and a to-be-welded workpiece pose parameter by utilizing the three-dimensional welding groove graph and the welding gun pose data, so as to determine a welding gun pose planning value; filling welding of the current pass is conducted along the welding groove according to the welding gun pose planning value, meanwhile, the welding gun pose real-time value is detected in real time, and the welding gun pose real-time value is corrected; in the return stroke, welding gun pose parameters are re-planned; and after multi-pass filling welding is completed, a welding gun is controlled to conduct cosmetic welding on the welding groove, and pipeline outside welding is completed. Therefore, the method can obtain a better welding effect.

A collaborative seam tracking method based on laser line scanning sensing includes: scanning, by using a laser line scanning sensor, a weld seam to obtain an original sampling deviation value; coordinate converting the original sampling deviation value to obtain a converted sampling deviation value, and then obtaining a first converted sampling deviation coordinate based on a current position coordinate of a motion mechanism; in response to an absolute difference between the converted sampling deviation value and a position value of a corresponding projection point on a preset path is less than a processing range of a galvanometer, performing a seam tracking; in response to the absolute difference between the converted sampling deviation value and the position value of the corresponding projection point on the taught path is greater than or equal to the processing range, performing Hermite interpolation, thereby obtaining an interpolation function as a motion trajectory.

The invention relates to the technical field of optical nondestructive testing equipment for the sealing performance of a blueberry bottle opening, in particular to an optical nondestructive testing device and method for the sealing performance of the blueberry bottle opening. According to the technical scheme, the optical nondestructive testing device for the sealing performance of the blueberry bottle opening comprises an optical testing device, mounting grooves, a lead screw, a mounting support, a mounting ring, a Fresnel lens, a micrometer caliper, a double-color laser line light source and a rotating motor, and the mounting grooves are formed in the two sides of the interior of the optical testing device; compared with a traditional optical nondestructive testing device for the sealing performance of the blueberry bottle opening, a single light source or a simple lens group is mostly adopted, a 360-degree annular area of the bottle opening is difficult to cover, illumination on the edge of the bottle opening or a complex curved surface area is insufficient, and tiny defects are prone to being omitted; according to the optical nondestructive testing device for the sealing performance of the blueberry bottle opening, the double-color laser line light source is matched with the Fresnel lens array, multispectral reflection data are synchronously collected, the surface micro-mark contrast is enhanced by utilizing wavelength difference, and the monochromatic light detection sensitivity is effectively improved.

According to the metal plate self-adaptive straightening method and technology based on the line laser contourgraph, the laser scanning contourgraph is arranged above a feeding part of a straightening machine, a line laser transmitter and an image sensor are arranged in the laser scanning contourgraph, and the line laser transmitter projects laser lines to a metal plate; the image sensor is used for capturing a plate image with a laser reflection light strip to assist the straightening control system to adjust the pressing roller, straightening of the metal plate can be adjusted in real time, and the straightening precision is improved; image data collected by an image sensor is transmitted to a data processing module, and the data processing module analyzes an image to obtain actual contour data of a plate before straightening; the actual contour data are compared with a preset ideal straightened plate contour model, the deviation value of each point is calculated, and a deviation matrix is generated; the straightening machine control system controls the position conditions of all the pressing rollers independently according to the deviation matrix, the accuracy of plate straightening can be greatly optimized, and continuous adjustment can be achieved.

Disclosed in the present invention is an apparatus for measuring the thickness of heat insulation cotton after hot pressing. The apparatus comprises an inspection platform, and a conveyor belt used for conveying heat insulation cotton, wherein the conveyor belt closely passes over the top plane of the inspection platform. The apparatus further comprises: a thickness measurement assembly, which is used for measuring the thickness of the heat insulation cotton when the conveyor belt stops; and a surface inspection assembly, which comprises a laser scanning portion and an image collection portion, wherein the laser scanning portion obliquely emits a laser line to the surface of the heat insulation cotton and moves along the surface of the heat insulation cotton, and the image collection portion is located right above the heat insulation cotton and performs image collection on the trajectory of the laser line. The present invention can measure the thickness of heat insulation cotton by means of a thickness measurement assembly, and can further perform image collection on the trajectory of a laser line by means of an image collection portion, and therefore the surface quality of the heat insulation cotton can be inspected on the basis of a change in the laser line in an image, and then in combination with the measurement of the thickness of the heat insulation cotton, the range of heat insulation cotton inspection can be wider, and inspection results are also more objective.

The invention discloses a vision and line structured light fused underwater three-dimensional measurement method. The method comprises the following steps: step 1, preprocessing an acquired image; 2, system parameter calibration is carried out on the preprocessed in-air color calibration image and the preprocessed in-air center line structure light calibration image, and internal and external parameters of a camera and a line structure light plane equation are obtained; 3, establishing a three-dimensional refraction model for the underwater color three-dimensional measurement object image and the underwater line structure light measurement object image, and performing affine transformation of the three-dimensional refraction model; 4, performing laser line center extraction on the transformed underwater line structured light measurement object image, and substituting the laser line center into the calibrated line structured light plane equation to perform point cloud reconstruction; and step 5, performing color processing on the underwater color three-dimensional measurement object image, and mapping the image to three-dimensional point cloud data. According to the invention, through affine transformation of the multi-medium refraction image and point cloud mapping of color multi-channel equalization, the precision and visual effect of underwater three-dimensional measurement are significantly improved.

The invention belongs to the field of pouring liquid level detection, and particularly relates to a vacuum pouring liquid level visual detection device and method based on a laser light source, and the device comprises an image collection unit which is arranged right above a pouring unit, is connected with an image processing host through a host connection cable, and is used for receiving a shooting instruction; the collected liquid level image of the pouring unit irradiated by the laser is sent to the image processing host for processing; the laser unit is connected with the image processing host and is used for emitting laser to the pouring unit; the pouring unit is connected with the pouring material port controller and used for receiving a pouring instruction of the image processing host and completing the pouring process according to an automatic judgment rule. And the image processing host is used for carrying out image processing operation on the liquid level image of the pouring unit irradiated by the laser and measuring the liquid level height through an image laser line position detection algorithm according to the obtained processed image. According to the invention, the accuracy degree of liquid level height calculation is greatly improved, so that the control accuracy of the automatic visual gating system is improved.

The invention relates to a vehicle detection method, device and system, and a computer readable storage medium, and relates to the field of vehicle security check. The vehicle detection method comprises the following steps: in a process that at least one to-be-detected vehicle runs on a detection road section, acquiring original point cloud data by using a multi-line laser radar, and determining a laser scanning line to which each point cloud of each to-be-detected vehicle in the original point cloud data belongs; according to the original point cloud data, obtaining vehicle point cloud data of each to-be-detected vehicle in the at least one to-be-detected vehicle; according to the vehicle point cloud data, determining at least one of position information and vehicle speed information of each to-be-detected vehicle; determining target point cloud data of each to-be-detected vehicle obtained by scanning at least one laser line of which the vertical field angle is greater than a vertical field angle threshold value from the vehicle point cloud data; and according to the target point cloud data of each to-be-detected vehicle, determining position information of a loaded object of each to-be-detected vehicle.

The invention provides a steel plate surface defect detection method and system based on multi-source data fusion, and relates to the technical field of steel production automatic detection.The method comprises the steps that water stain pretreatment is conducted on the surface of a steel plate, the residual humidity is reduced to be below a preset low humidity threshold value, and a dry surface is obtained; based on the dry surface, collecting multi-source data through a 3D laser line scanning camera and a 2D industrial line scanning camera which are synchronously triggered, obtaining line scanning laser point cloud data, a line scanning reflectivity gray level image, a line scanning depth gray level image and an area array reflectivity color image, and establishing a space corresponding relation among the multi-source data; and forming a multi-source data set based on the spatial correspondence among the multi-source data. According to the method, a high-precision and high-robustness steel plate surface defect detection system covering the whole detection process is constructed by eliminating water stain interference, unifying a multi-source data reference, optimizing data quality, accurately identifying defects and realizing automatic feedback and data closed loop.

The invention relates to the technical field of optical measurement and metering equipment, in particular to an on-line precision calibration method for a laser line module in a high-low temperature environment, which comprises the following steps: acquiring horizontal and vertical offsets of a projection light spot of the laser line module, and collecting the temperature of the position of a laser transmitter; respectively calculating a horizontal offset interference degree and a vertical offset interference degree at each acquisition moment, and further obtaining a horizontal interference ratio and a vertical interference ratio; according to the change trend of the horizontal offset interference degree and the vertical offset interference degree within the preset duration before each acquisition moment, the horizontal and vertical interference variation of each acquisition moment is obtained, and then the compensation regulation factors of the horizontal offset and the vertical offset of each acquisition moment are obtained, and the horizontal offset and the vertical offset are compensated. And compensation correction values of the horizontal offset and the vertical offset are obtained, so that the projection precision of the laser line module is calibrated. According to the invention, the precision of projection calibration of the laser line module can be improved.

The invention relates to the technical field of 3D vision measurement, and discloses a 3D binocular line laser weld contour data acquisition method and system. Two core steps of obtaining surface reflection characteristic information of a to-be-scanned area and adjusting output intensity of laser rays projected by a line laser device at different positions of the to-be-scanned area according to the surface reflection characteristic information are introduced, so that the problem that when a high-reflection welding seam is processed by a traditional method, the laser rays cannot be projected by the line laser device is effectively solved; the problem of image overexposure or underexposure caused by dramatic change of illumination characteristics is solved.

The present invention discloses a flatness detection method for a tower flange of a wind turbine generator set, comprising two steps: a flange plane 3D point cloud acquisition stage and a flatness calculation stage; wherein: the flange plane 3D point cloud acquisition stage comprises the following steps: rotation control comprises the following steps: the 3D vision detection module is accurately positioned; the 3D vision detection module performs laser line scanning on the annular surface to be detected by the flange; the 3D vision detection module performs 3D point cloud acquisition on the flange surface to be detected; the flatness calculation stage comprises the following steps: flatness calculation is performed according to the 3D points output by the 3D vision detection module; and the result is output; compared with the prior art, the flatness detection method for the tower flange of a wind turbine generator set of the present invention integrates motion control and 3D vision detection technology, provides an automated detection method, innovates the current single instrument manual detection method, and improves the efficiency and speed of tower flange flatness detection.

The invention discloses an anti-interference target laser ray extraction method. The method comprises the following steps of S1, performing Gaussian filtering processing on an image; s2, determining possible starting and ending position points of the target laser ray: the starting and ending position points of the target laser ray are respectively b and b1, the difference value of two adjacent pixel gray values is diffval, and the determination method of b and b1 is as follows: the determination method of b: b-a < = diffval and c-b > = diffval; the determination method of b1 is as follows: b1-a1 is greater than or equal to-diffval, and b1-c1 is less than or equal to-diffval; wherein a and c are two adjacent points on the two sides of b in the width direction of the laser line respectively, and a1 and c1 are two adjacent points on the two sides of b1 in the width direction of the laser line respectively; s3, setting the line width of a target laser line; s4, extracting the center point of the target laser ray: extracting the center of the laser ray by using a gray gravity center method; s5, extracting a target laser ray; the method can be applied to the 3D laser contourgraph to extract correct laser rays under the condition of interference, and serious loss of industrial production caused by inaccurate extraction of the laser rays is avoided. According to the invention, the laser line extraction precision is improved.

The invention discloses a fuel cell stack core straightness and perpendicularity detection device and method. A reference surface is positioned through a visual camera, and a transmission mechanism drives a laser line scanning instrument to perform line scanning point taking on an upper end plate and fit a reference plane; moving a visual camera through an X-axis or Y-axis transmission mechanism, scanning a long-edge extension line and a short-edge extension line of the upper end plate on a reference surface of the upper end plate along the horizontal direction, obtaining an intersection point, determining a coordinate origin, and performing reference determination by taking the long-edge extension line and the short-edge extension line as X-direction or Y-direction reference of a coordinate system; the transmission mechanism drives the laser line scanning instrument to carry out line scanning on the straightness of the side edge of the reactor core assembly, straightness line scanning data are obtained, line scanning point position coordinate values are obtained, data analysis is carried out according to the coordinate values, and compared with the prior art, high-precision detection of the straightness and perpendicularity of the fuel cell stack core is achieved, and the detection precision is improved. And the assembly quality and performance of the reactor core assembly are ensured.

The invention discloses laser measurement equipment for indoor decoration construction and a method thereof, the laser measurement equipment comprises a transmission seat, a laser horizontal measurement instrument body and a glass protection plate, and the output end of the transmission seat is fixedly connected with the bottom of the laser horizontal measurement instrument body. By arranging the transmission part, when a user starts the transmission seat to drive the laser horizontal measuring instrument body to rotate and adjust clockwise, the right side of the transmission part rotates towards the top, and then the transmission part continuously drives the cleaning cotton to clean the front surface of the glass protection plate, so that the problem that a large amount of construction dust is generated during decoration is solved; the problem that the laser measurement effect of the laser horizontal measurement instrument is affected due to the fact that the laser lines are interrupted because a part of laser emitted by the laser horizontal measurement instrument is blocked by dust when the dust falls on a glass protection plate at the laser emission position of the laser horizontal measurement instrument is solved. And the effect of cleaning the glass protection plate during adjustment is achieved.

The invention relates to the technical field of joint robot calibration, in particular to a robot motion indication light spot position rapid acquisition device based on four PSDs, and aims to solve the problem that in an existing joint robot calibration device based on the PSDs, it is difficult to adjust a robot to be aligned to the central points of the PSDs. The second PSD is fixed above the first laser reflection structure, the third PSD is fixed below the second laser reflection structure, and the fourth PSD is fixed above the second laser reflection structure. The first laser reflection structure and the second laser reflection structure are fixed to the reference plate, the reference plate is connected with the lower shell through bolts, a power source, a signal processor and a wireless transmitter are fixed in the lower shell, the signal processor and the wireless transmitter are connected with the power source, and the wireless transmitter and the wireless data transmission radio station conduct transmission through wireless communication. And the wireless data radio station is connected with the upper computer. The problem that laser ray constraint must be aligned with the center of the PSD for repeated adjustment is solved, the operation difficulty is reduced, and the calibration efficiency and precision are improved.

The invention relates to the technical field of metal composite plate preparation, and provides a metal composite plate preparation method which comprises the steps that a first metal plate and a second metal plate are stacked, and a first to-be-welded interface is formed by the first metal plate and the second metal plate; performing high-frequency laser line scanning on the first to-be-welded interface in the width direction of the first metal plate to form a line molten pool, and pressing the line molten pool; and along the length direction of the first to-be-welded interface, high-frequency laser line scanning is carried out for many times to form a plurality of line molten pools, and the line molten pools are pressed, so that the first metal plate and the second metal plate at the first to-be-welded interface are combined. According to the preparation method of the metal composite plate, the metal plate can be continuously welded to form the metal composite plate by performing high-frequency laser line scanning on the first to-be-welded interface edge and pressing the edge-to-line molten pool, and the interface of the metal composite plate is not prone to cracking, warping and the like, so that the preparation efficiency of the metal composite plate is improved, and the production cost is reduced. And meanwhile, the preparation quality of the metal composite plate is improved.

The invention relates to the technical field of three-dimensional measurement, in particular to a linear structured light bearing bush and gear three-dimensional measurement method and system, and the method comprises the following steps: 1, data collection: employing a CCD camera arranged on a measurement platform to shoot laser images containing laser line patterns from different angles; step 2, image processing: extracting feature points in the laser image by using a computer vision algorithm; and 3, three-dimensional reconstruction: based on the known CCD camera parameters and the position change of the feature points on the to-be-measured object, obtaining three-dimensional data of the to-be-measured object by using a space coordinate conversion technology. And step 4, error correction: carrying out error analysis on the obtained three-dimensional data. And step 5, result output: outputting final three-dimensional measurement data of the to-be-measured object based on the corrected result. Through the automatic and intelligent measurement means, the measurement period can be further shortened, and the production efficiency is improved.

The invention discloses a depth and color fused true color three-dimensional point cloud acquisition method and system, and the method comprises the steps: setting a plurality of lasers to emit specific spectrum laser, enabling the specific spectrum laser to be reflected by the surface of an object, and filtering the specific spectrum laser through a narrow-band optical filter of a corresponding wave band, thereby enabling a corresponding sensor to only receive a single-spectrum laser line emitted by a corresponding laser emitter. A laser center point position and an intensity response value are extracted through a sub-pixel-level laser line refinement technology; calibrating the unified measurement coordinate system through sensor pose calibration; according to the light transmittance of an optical filter, spectral response characteristics of an image sensor and the like, correcting and compensating the intensity response value, calculating a compensation factor to obtain a color value of a measurement point, fusing a three-dimensional coordinate with corrected color information, outputting true color point cloud data, and presenting depth and color information of an object at the same time. The three-dimensional information and the color information of the object are synchronously obtained, so that the measurement data contain rich spectral characteristics.

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 laser digital display tape provided by the present invention comprises a tape core seat, a laser ranging module, a calibration detection unit and a corner measurement unit which are installed in a shell, one side of the tape core seat is fixedly connected with a roller, and the other side of the roller is connected with a tape core cover; the center of one side of the ruler core cover is fixedly connected with a rotating shaft, and one end of the rotating shaft is connected with a corner measuring unit; a ruler strip is tightly wound on the outer side of the roller through a ruler spring, and black and white stripes are printed on the ruler strip in a silk-screen mode. The measuring tape is compact in overall structure, does not occupy space, is integrated with the laser ranging module, solves the problem that the measuring length of the measuring tape is limited, can achieve high-precision distance measurement within the range of dozens of meters or even hundreds of meters, enables a laser line to be close to the tape strip as much as possible, brings convenience to a user for measuring the same target object through the measuring tape and laser, and improves the measuring efficiency. The use convenience of a common measuring tape is improved, direct reading by a user is facilitated, the measuring distance is prolonged, and long-distance linear measurement is realized.

The invention provides a radiation therapy laser positioning system performance test method, which can be applied to the technical field of medical equipment, and comprises the following steps: selecting measurement points on a laser line of a laser positioning system, and measuring the position coordinate of each measurement point; on the basis of the position coordinates of the measuring points, a fitting straight line in the X-axis direction, a fitting straight line in the Y-axis direction and a fitting straight line in the Z-axis direction of the laser positioning system are obtained; the straightness of a laser line of the laser positioning system and the perpendicularity of a laser cross positioning line are obtained through calculation; and measuring the distance deviation between the reference point determined by the laser positioning system and the reference point of the radiotherapy system to obtain a consistency test result of the reference points of the laser positioning system and the radiotherapy system. On the premise that the existing test requirements are met, the straightness of the laser rays, the perpendicularity of the laser cross curve and the positioning accuracy of the radiotherapy positioning laser system are improved. The invention also provides a radiation therapy laser positioning system performance test system and electronic equipment.

The utility model discloses a magnetic leveling device of a pendulum bob type laser demarcation device, which comprises a hanging bracket and a pendulum body hung in the hanging bracket, a magnetic damping body consisting of a plurality of magnets is arranged on the hanging bracket according to a Halbach array, and a magnetic field which is made of non-magnetic conductive materials and can cut magnetic lines in a magnetic field generated by the magnetic damping body is arranged on the pendulum body. The utility model aims to provide a damping piece which adopts two different forms of Halbach arrays to generate magnetic damping, and compared with the existing common magnet arrangement scheme, the magnetic flux is obviously enhanced, so that a stronger magnetic damping effect is generated, and the magnetic damping effect is improved. The magnetic leveling device can effectively prevent the swinging body from swinging excessively, so that the leveling time is effectively shortened, the leveling speed and precision of the laser demarcation device are further improved, and the working efficiency is remarkably improved.

The invention discloses a plug-and-play Rydberg atom preparation device based on a three-photon absorption method, which comprises a laser generator, a laser input module, a laser detector, three frequency stabilization modules and a temperature and humidity control module, and is characterized in that the laser generator is used for preparing laser to be measured; the input module is used for transmitting and introducing laser; the saturated absorption frequency stabilization module, the electromagnetic induction transparent frequency stabilization module and the heating electromagnetic induction transparent frequency stabilization module all adopt a glass bonding technology to build frequency stabilization light paths; the laser detection device is used for detecting laser lines in the saturation absorption frequency stabilization module, the electromagnetic induction transparent frequency stabilization module and the heating electromagnetic induction transparent frequency stabilization module and locking the laser frequency corresponding to the specific wavelength. According to the invention, a frequency-stabilized light path is built by using a glass bonding technology, so that devices and the size are greatly reduced, and the integration level of a light path system is greatly improved; moreover, the degree of freedom of device adjustment can be reduced, and the reliability and stability of a laser frequency stabilization optical path are effectively improved.

The present invention provides a calibration device and method for a 3D camera, and relates to the technical field of camera calibration. The calibration device for a 3D camera includes a calibration plate, a translation stage, and a linear structured light laser mounted on the translation stage, wherein the translation stage is used to carry a 3D camera, and the translation stage can be moved closer to or farther from the calibration plate, and a detachable groove structure is provided on the surface of the calibration plate facing the translation stage, and a groove is provided on the groove structure. The cross-sectional contour of the groove on the calibration plate is measured by triangulation using a laser line, the coordinates of the points on the cross-sectional contour are extracted, and a set of equations is established with the coordinates of the known edge points of the groove, so that the coordinate transformation matrix coefficients can be solved. According to the coordinate transformation matrix coefficients, the corresponding object coordinates can be found according to the image coordinates, which not only has high solution efficiency, but also reduces the influence of camera lens distortion and has high calibration accuracy.

The invention discloses a belt tearing detection method based on a multi-task deep learning network, and the method comprises the steps: firstly projecting laser to the surface of a belt, and then collecting the image data of the belt in real time through a camera; carrying out augmentation and preprocessing on the image data, and adding a light strip image with noise at the same time; extracting a laser stripe center point as a data label by using an improved gray gravity center algorithm; a U-Net network is improved, a Dropout layer is added after coding and decoding of each layer, then different weights are set for network training, and an improved MU-Net model is obtained; fitting the center point of the laser line output by the full connection layer by using a linear difference method to obtain a light stripe center line, and taking the light stripe center line as a final output result of the stripe center line; and analyzing the center lines of the stripes, and identifying different defects of belt tearing by judging the distance between two adjacent pixel points and the included angle between the tangent lines of the adjacent pixels. According to the invention, real-time monitoring of the state of the belt is realized, and powerful support is provided for preventive maintenance of industrial equipment.

The invention relates to the technical field of underwater operation equipment, solves the problems that existing dam underwater crack repairing depends on divers or complex equipment, the automation degree is low, the construction period is long, the comprehensive cost is high and the adaptability is poor, and particularly relates to concrete dam underwater crack repairing and leaking stoppage equipment which comprises a robot for underwater cruise, and the robot comprises a machine body; the cruise vision module comprises a foresight navigation lens group, a look-around lens group, a bottom wide-angle lens, an active illumination system, an obstacle avoidance sonar system and a topographic mapping sonar; the crack detection module comprises a crack scanning ultrasonic detector, a high-resolution strip-shaped scanning lens group, a laser ray scanner, a multispectral imager, a laser distance measuring sensor, a water flow sensor and an attitude sensor, the crack repairing module comprises two groups of mechanical arms for punching and grouting, and the central controller controls the robot to execute. Crack filling, structure bonding recovery and long-acting sealing are achieved, and meanwhile the construction difficulty and the comprehensive cost are reduced.

Provided is a handheld device and a method for detecting weld defects on a weld surface. The handheld device projects a thin beam laser light on the weld surface using a laser emitter. A processor receives the projected thin beam laser light from a laser receiving sensor, which indicates captured contours of the weld surface. The captured contours are processed to filter laser lines and extract contours. The extracted contours are utilized to classify a type of weld surface such as, surface joints, butt joints, corner joints, and tee joints. The processor, upon classifying, automatically adjusts angle and position of the laser light to be projected on the weld surface and detects deviation patterns by employing an ensemble of a laser line extraction process and a pixel width measurement process. The processor determines the detected deviation patterns as one or more defects on the weld surface that is being inspected.