1573 results about "Visual positioning" patented technology

The invention relates to the technical field of laser engraving, and discloses a laser engraving method and system capable of automatically correcting coordinates of a galvanometer and a camera, and the method comprises the following steps: calculating a target conversion coefficient based on a calibration size and a pixel size of an image collected by the camera, and controlling the galvanometer to perform laser etching on a cross mark on the surface of a PCB (Printed Circuit Board), recording origin data of a galvanometer coordinate system; adjusting the position of the camera to enable the cross center of the view of the camera to coincide with the cross mark, and calculating the coordinate offset; coordinate space transformation is carried out on all the points to be machined, and galvanometer marking coordinate data are obtained; after laser engraving is carried out on the galvanometer marking coordinate data, an engraving area image is collected, the position residual error and the size proportion deviation are calculated, laser engraving is carried out again, and a laser engraving result is obtained. The technical problem that the visual positioning coordinate and the galvanometer marking coordinate in the laser engraving system are not consistent is effectively solved.

The invention relates to the technical field of unmanned aerial vehicle navigation and control, and discloses an unmanned aerial vehicle indoor and outdoor seamless navigation method fusing Beidou and visual positioning. Comprising the following steps: S1, synchronously acquiring original observation data of a Beidou satellite navigation system of an unmanned aerial vehicle, an image sequence acquired by a visual sensor and inertial data of an inertial measurement unit; s2, processing the original observation data of the Beidou satellite navigation system to obtain the position of the unmanned aerial vehicle, according to the unmanned aerial vehicle indoor and outdoor seamless navigation method and system fusing Beidou and visual positioning, through a self-adaptive fusion filter module, the fusion weight of the unmanned aerial vehicle and visual positioning is dynamically adjusted according to the Beidou signal quality; smooth transition of indoor and outdoor navigation main sources is achieved, pose jump is effectively avoided, a tight coupling depth fusion algorithm is adopted, the precision and robustness of the system are improved, the continuous and stable flight capacity of the unmanned aerial vehicle in complex indoor and outdoor environments is ensured, and the application scene is expanded.

The invention provides a pathological image visual positioning method and system, equipment and a storage medium, and belongs to the technical field of image recognition, and the method comprises the steps: extracting visual features based on a target pathological image, and determining a semantic feature vector and a knowledge feature vector based on first text description; the target pathological image is a pathological image to be subjected to target area positioning, and the knowledge feature vector is used for representing knowledge information associated with the content of the target pathological image; fusing the semantic feature vector and the knowledge feature vector to obtain a fused text feature; performing cross-modal fusion on the fused text features and the visual features to obtain fused multi-modal features, and obtaining fusion representation based on the fused multi-modal features; and based on the fusion representation, positioning a target area in the target pathological image through a multi-layer perceptron to obtain position information of a bounding box of the target area. The method can improve the capability of accurately and flexibly positioning the pathological image region level.

The invention relates to a tin ring automatic focusing laser soldering tin fixing and 3D welding spot detecting system, and belongs to the technical field of industrial automatic laser processing. The system is characterized in that a tin ring preparation module identifies pins in real time and dynamically optimizes winding parameters through visual positioning and a self-adaptive winding head, and performs defect identification and pre-repair analysis through multispectral imaging; the soldering tin positioning module is used for switching stations through a rotary table, and precise positioning and clamping of a product are realized by combining laser contour sensing and a focusing compensation algorithm; the laser tin soldering module can visually identify tin materials and automatically call welding parameters, laser power and action time are regulated and controlled in real time by establishing a regional heat conduction model, and gradient cooling is implemented after welding; and the 3D detection module adopts dual-mode scanning, constructs a welding spot three-dimensional model based on fusion data, recognizes microcracks through super-resolution processing, and discriminates the welding spot quality. And automation and intellectualization of the whole process from tin ring preparation to welding spot quality detection are achieved.

The invention discloses a multi-modal multi-scale retrieval enhancement generation method, system and device applied to external knowledge questions and answers and a medium. The method comprises question perception, multi-modal multi-scale query fusion coding, dense recall and answer generation. Analyzing a key query phrase from the question through a fine-tuned instruction language model, and accurately positioning a region of interest corresponding to the phrase in an image by using an open set visual positioning model; multi-source information is compressed and distilled into an optimal query vector through a deep fusion network integrating multi-head self-attention and an information bottleneck theory; executing a maximum inner product search to recall related knowledge; guiding the large language model to synthesize all information to generate a final answer; the system, the equipment and the medium directly perform feature fusion in the vector space based on the method, so that challenges such as information loss and cascading errors caused by a traditional normal form can be effectively dealt with, high correlation and high accuracy of retrieval knowledge are ensured, and accurate and reliable image-text questions and answers are realized.

The invention belongs to the technical field of unmanned aerial vehicles, and provides an omni-directional vision positioning system and method and an unmanned aerial vehicle, the omni-directional vision positioning system is installed on a fuselage of the unmanned aerial vehicle, and the omni-directional vision positioning system comprises an image acquisition device used for acquiring a real-time scene image in an omni-directional range in the flight process of the unmanned aerial vehicle; the inertial measurement sensor is connected with the image acquisition equipment and is used for synchronously acquiring motion state data of the unmanned aerial vehicle; the navigation system is connected with the image acquisition equipment and the inertial measurement sensor and is configured to execute at least one of distortion correction processing and feature extraction processing on the real-time scene image to obtain a processed image; and according to the motion state data and the processed image, determining current pose information of the unmanned aerial vehicle in a space coordinate system constructed by taking a take-off point of the unmanned aerial vehicle as an original point. The positioning precision and robustness of the unmanned aerial vehicle in a complex environment are improved, and the requirement of high-precision autonomous flight is met.

The invention discloses a multi-modal remote sensing visual positioning method and device based on scene knowledge enhancement and a medium, and relates to the technical field of remote sensing visual positioning. The method comprises the following steps: firstly, preprocessing a plurality of remote sensing images, generating scene knowledge enhanced text description, forming a visual positioning data set, and dividing the visual positioning data set into a training set, a verification set and a test set; constructing a visual positioning model, and obtaining an optimal model through training, verification and testing; inputting a to-be-queried text to obtain remote sensing image coordinates. According to the method, cross-modal fusion of knowledge enhancement is realized, scene knowledge is embedded into a visual positioning framework, and the problem of inference of implicit semantics in a remote sensing scene is solved; multi-scale image features and scene knowledge are fused layer by layer through multi-round cross-modal attention iteration, and semantic understanding from coarse granularity to fine granularity is achieved; a similarity threshold is introduced to screen a high-correlation image region, and background interference is reduced in combination with loss constraints. The LLaMA2 is subjected to efficient fine tuning in combination with the LoRA technology, end-to-end coordinate generation is supported, and both performance and calculation efficiency are considered.

The invention relates to the technical field of image processing, in particular to an image local enhancement super-resolution method based on text prompt. According to the technical scheme, the method comprises the steps of obtaining a low-resolution image and a text prompt provided by a user; and inputting the low-resolution image and the text prompt into a visual positioning model to generate a region-of-interest mask which is used for identifying the position of a target region corresponding to the text prompt in the image. According to the invention, the visual positioning model is driven to automatically identify the region of interest of the image through text prompt, fine reconstruction branches are configured for the region of interest, lightweight reconstruction branches are configured for the background, and global visual consistency is guaranteed in combination with the fusion module, so that the readability and the identifiability of details of the region of interest are remarkably improved; the method is advantaged in that accurate requirements of intelligent monitoring, document OCR, medical image and other scenes are satisfied, calculation resource distribution is substantially optimized, calculation power waste of a background area is avoided, and image super-resolution efficiency and practicality are improved.

The invention relates to the field of robot vision positioning, and discloses an intelligent hand-eye calibration and self-adaptive correction system and method. According to the method, a mechanical arm is controlled to drive a camera to collect multi-modal calibration data, a convolutional neural network is utilized to identify a calibration plate mark point, and a three-dimensional coordinate under a camera coordinate system is calculated in combination with depth information; meanwhile, on the basis of an encoder and torque data, flexible deformation of the mechanical arm is compensated through a kinematic model and a self-adaptive rigidity model, and three-dimensional coordinates under a base coordinate system are obtained. And obtaining a hand-eye transformation matrix by solving a transformation relation between the two point sets. And repeatedly calibrating before and after operation, inputting the difference of the two transformation matrixes into the fault diagnosis neural network, outputting fault type probability distribution, and generating a maintenance strategy. According to the invention, high-precision hand-eye calibration, automatic compensation of flexible deformation and intelligent diagnosis of system state change can be realized, so that the long-term precision and reliability of a visual positioning system are improved.

According to the multi-source information fusion robot positioning method and system for the unstructured environment provided by the invention, fusion processing is carried out on multi-source information data such as laser point cloud, environment image information and acceleration which are acquired in real time, so that the multi-source information fusion robot positioning method and system for the unstructured environment can be realized in a severe weather state and a dynamic unstructured environment. Carrying out high-robustness real-time positioning under the condition of violent illumination change; in a geometrically degraded roadway area, registration positioning of laser point cloud emission fails due to lack of enough external characteristics, and at the moment, the visual positioning module can still work normally and complete robot repositioning by detecting matched image information; in an environment lacking enough illumination, the laser point cloud and acceleration information can still provide enough positioning fusion data input for the computing unit.

The invention relates to the technical field of mechanical arms, in particular to a chemical emergency disposal-oriented manipulator cross-modal sensing control system which comprises an environment scanning module, a visual positioning module, an obstacle avoidance planning module, a pose compensation module, a decision mapping module and a grabbing updating module. Obtaining environment state information; an operation view angle image of the manipulator is collected, visual features of the operation view angle image are solved, and pose information is obtained; obstacle avoidance correction is conducted on the motion trail of the mechanical arm, and a collision-free approaching path is obtained; the operation view angle image and the multi-mode micro-touch signal are fused, error compensation is carried out on the pose information, and an accurate grabbing pose is obtained; mapping object surface physical attributes sensed by the multi-mode micro-tactile signals to obtain a self-adaptive grabbing instruction; the manipulator is controlled to conduct grabbing through the self-adaptive grabbing instruction, and environment state information is updated; according to the invention, the accuracy of cross-modal sensing control of the manipulator can be improved.

The invention discloses a UWB navigation system and method based on visual feature fusion, the system comprises a UWB positioning module, a visual perception module, a data fusion processing module and a navigation control module, the UWB positioning module adaptively generates UWB anchor point layout, calculates UWB coordinates in real time, and then obtains UWB positioning data according to fruit tree coverage rate and soil humidity calibration; the visual perception module collects orchard images to recognize fruit trees and obstacles, obtains visual coordinates and corrects the visual coordinates to obtain visual positioning data; the data fusion processing module calculates weights of the UWB positioning data and the visual positioning data, fuses the UWB positioning data and the visual positioning data according to the weights to obtain fusion coordinates, and performs error correction to obtain optimal coordinates; and the navigation control module calculates a target path point according to the optimal coordinate, the obstacle recognition result and the weeding operation path, and controls the weeding robot to run to the target path point. According to the invention, high-precision navigation of the weeding robot in a complex orchard environment is realized.

The invention provides a visual positioning and grabbing force control method for a palletizing robot in the intelligent manufacturing industry, and relates to the technical field of intelligent manufacturing. The multispectral visual positioning module at least comprises a visible light imaging unit, a near-infrared imaging unit and a structured light imaging unit; s2, on the basis of the material three-dimensional information, fitting a material motion track through a motion trend prediction model, and calculating a target grabbing position in advance; through visible light + near infrared + structured light multispectral fusion, the problem of interference of dust and haze on visual positioning is solved, and the positioning error is reduced to be within 1mm in combination with an adaptive weight algorithm; and meanwhile, the grabbing position is calculated in advance through the motion trend prediction model, the dynamic positioning response speed is increased, compared with the prior art, the performance is higher, and the high-speed dynamic stacking requirement is met.

The invention relates to the technical field of image processing and target tracking, in particular to a cross-view fusion multi-person detection and tracking method for outdoor view variety, which comprises the following steps of: reconstructing a multi-view picture into three-dimensional / BEV space priori under a unified world coordinate, and feeding back the priori to each camera position feature by using space enhancement attention to obtain a consistent and steady multi-person detection result; then, carrying out online iteration updating on the three-dimensional position and posture of the figure in a tracking mode, introducing OKS-3D gating to combine the 2D key point consistency with the BEV spatial distance, and keeping a uniform ID and a continuous track across cameras; the method solves the problem that the prior art cannot effectively meet the requirements of real-time, accurate and stable target detection, tracking and identity association of an outdoor variety in a complex shooting environment, realizes time-space structured output of accurate detection, stable tracking, unified ID and visual positioning, can directly serve for director scheduling and post-production, and is high in practicability. And the overall manufacturing level and quality of the exterior variety are improved.

The invention relates to the technical field of crossing of artificial intelligence and intelligent manufacturing, discloses a position self-adaptive AI regulation and control method and system for continuous heat treatment of a fastener, and aims to solve the problems of non-uniform temperature field, high hardness dispersion, increase of quenching cracking rate and the like caused by lack of position perception and dynamic compensation lag in the prior art. According to the method, in-furnace coordinates of fasteners are tracked in real time through a high-speed visual positioning device, a thermal demand prediction neural network model with space-time convolution mixed with a gating circulation unit is constructed in combination with the conveying speed, thermal environment parameters and material characteristics, and an instantaneous thermal power instruction is output; the system comprises a high-speed visual positioning module, a motion state monitoring module, a thermal environment sensing module, a thermal demand prediction module and the like, and drives a high-density partitioned infrared heating array to realize millisecond-level local thermal compensation. According to the method, the temperature deviation of the junction of the temperature zones can be controlled within + / -8 DEG C, the hardness dispersion and the cracking rate are remarkably reduced, and the heat treatment consistency and the finished product yield are improved.

The invention belongs to the technical field of visual positioning, and relates to a visual positioning method based on semantic comprehension and attribute distinguishing enhancement. The framework mainly comprises a feature coding module, a semantic sensitive data enhancement module, a fine-grained attribute guiding module and a multi-stage cross-modal decoder module. Specifically, the feature coding module performs feature coding on an input graph. The semantic sensitive data enhancement module generates a plurality of queries which are consistent with long text semantics by keeping the consistency of spatial relation words in combination with a large language model, so that a training data set for the long text is expanded. The fine-grained attribute guiding module extracts attribute prior information from a text query in combination with a text graph model and an image encoder, constructs a visual feature representation with higher discrimination by using the information guiding model, and generates a target query with attribute difference at the same time.

The invention provides a visual-language model training method and device and related equipment, and the method comprises the steps: constructing a training data set comprising an image-text alignment task, a text-driven visual positioning task and a plain text reasoning task through determining a visual encoder and a heterogeneous language model which form a multi-modal heterogeneous recognition model; carrying out dimension dynamic alignment adaptation on the heterogeneous language model, enabling a parameter architecture of the heterogeneous language model to be matched with a visual feature dimension output by a visual encoder, and carrying out supervision and fine adjustment on the visual-language model by adopting a freezing-unfreezing two-stage training strategy based on a training data set; and performing parameter training on a cross-modal alignment module connected with the visual encoder and the heterogeneous language model so as to obtain a trained visual-language model. According to the training method, time consumed by model training is saved, and the convergence speed is increased. Through dimension dynamic alignment and hierarchical weight mapping, the pre-training language ability is reserved to the maximum extent, the plain text task performance loss is reduced, and disastrous forgetting of the model is avoided.

The invention provides a laser cutting production line task automatic allocation control method based on multi-device state coupling analysis, and belongs to the technical field of intelligent manufacturing and industrial automation. The method comprises the following steps: constructing a dynamic closed-loop control process through a central control scheduling system: receiving a task information packet of an MES; constructing an equipment state vector based on a real-time station state, and introducing a dynamic weight factor set to generate a weighted state vector; performing task triggering judgment through a coupling triggering judgment function in combination with the task dependency graph and historical task records; when the conditions are met, a task instruction is issued to the target station; and feeding back the state and updating the historical record after the task is completed. The invention further relates to AGV intelligent scheduling, visual positioning compensation, process parameter dynamic adjustment, predictive conflict detection, weight self-optimization and the like. According to the method, the production line cooperation efficiency is remarkably improved, manual intervention and system delay are reduced, and the method is suitable for an intelligent laser processing scene in which multiple devices run in parallel.

The invention discloses an automatic courtyard pool floating object salvage system based on image recognition, and relates to the technical field of intelligent robots, and the system comprises an image sequence acquisition module which is used for continuously capturing and transmitting an image sequence; the hydrological state sensing module is used for quantifying the dynamic and morphological characteristics of the water surface to obtain an average drift velocity vector and a water surface gradient field; the visual target positioning module is used for detecting a floating object to obtain an apparent two-dimensional coordinate vector; the trajectory fusion prediction module is used for fusing the average drift velocity vector, the water surface gradient field and the apparent two-dimensional coordinate vector, and calculating predicted physical coordinates; and the salvage operation execution module is used for receiving the predicted physical coordinates and driving the mechanical arm to complete salvage operation. According to the method, hydrological perception and visual positioning are fused, visual errors caused by dynamic disturbance of the water surface are corrected, the floating object track is predicted in combination with system delay, salvage operation is innovated from passive following to active interception, and the success rate and efficiency are remarkably improved.

The invention discloses an agricultural tractor visual positioning method and system, and belongs to the technical field of agricultural tractor visual positioning, and the method comprises the steps: extracting the multi-scale features of a farmland scene image, and carrying out the feature fusion, and obtaining the multi-scale fusion features; performing adaptive weighting on the multi-scale fusion feature map to generate a final feature map so as to obtain a feature point probability heat map and a feature point descriptor; extracting coordinates and detection confidence of the feature points according to the feature point probability heat map; when the detection confidence of the candidate feature points meets a confidence threshold requirement, matching the descriptors of the candidate feature points with the descriptors of the feature points of the adjacent frames in space to obtain matching pairs; and calculating according to the coordinates of the candidate feature points in the matched pairs, the coordinates of the feature points of the adjacent frames and the internal reference of the image acquisition equipment, and outputting the real-time pose of the tractor. According to the method, the spatial distribution of the feature points can be optimized, so that the discrimination is stronger, and the robustness and the positioning precision of the feature points in a farmland environment are improved.

The invention discloses a medical visual question and answer model construction method based on a feedback loop. The method comprises the following steps: step 1, generating initial candidates; step 2, performing multi-dimensional candidate evaluation; 3, iterative optimization of feedback driving is carried out; step 4; and 5, efficient fine adjustment of model parameters: performing fine adjustment on a pre-trained multi-modal large language model by using the enhanced training data set constructed in the step 4 to obtain a final medical visual question and answer model. According to the method, a feedback loop mechanism is introduced, and cooperative and iterative self-correction and optimization are carried out on generated text interpretation (Ratio) and a visual localization region of interest (RoI), so that the fact reliability, the logic consistency and the visual strong relevance of a final output result are systematically ensured.

The invention relates to the technical field of visual positioning, and discloses a visual positioning method, device, equipment and system. The method comprises the steps of obtaining a current image frame; extracting a first candidate frame similar to the current image frame from a map database; checking the continuity between the current image frame and the first candidate frame according to the current image frame and the first candidate frame to obtain a second candidate frame adjacent to or connected with the timestamp of the current image frame; obtaining an initial pose of the current image frame relative to the second candidate frame, and calculating a pose of the current image frame in a map according to the initial pose; checking whether the current image frame is a false detection frame or not according to the pose; and when the detection result is not the false detection frame, performing visual positioning according to the pose. According to the method and the device, the accuracy and the precision of visual positioning are improved, the retrieval efficiency is improved when the key frame is searched, the recall rate is relatively high for a mismatching condition, and the probability of missed detection is reduced.

The invention discloses a full-automatic packaging assembly line for semiconductor material trays, which comprises a material taking and feeding mechanism, a material tray label tearing and pasting mechanism, a moisture-proof monitoring mechanism, a bagging and packaging helium detection mechanism and a boxing and packaging redetection mechanism which are arranged in sequence, and the assembly line realizes material tray grabbing, visual detection and overturning, old label tearing and new label pasting; the full-process automatic operation comprises the following steps of: automatically placing a humidity card and a drying agent, bagging an aluminum foil bag, vacuumizing, filling helium, carrying out heat sealing, carrying out helium detection, and finally automatically boxing, sealing, rechecking, labeling and stacking. By integrating multi-stage visual positioning and detection, double-path parallel heat sealing and closed-loop quality control, the packaging efficiency and precision are remarkably improved, the sealing performance, moisture resistance and traceability of product packaging are ensured, and the automatic packaging machine is particularly suitable for automatic packaging of precise electronic components such as semiconductors.

The invention relates to the technical field of numerical control machining, in particular to a five-axis drilling numerical control machine tool based on visual positioning and a machining method, and the five-axis drilling numerical control machine tool comprises a machine tool body, a numerical control system, a contact type measuring head, a visual system and a calculation control unit. Coordinates of at least three non-collinear reference features on a workpiece in a machine tool coordinate system are automatically collected through a contact type measuring head, the deviation between the workpiece coordinate system and the machine tool coordinate system is automatically calculated through a coordinate system transformation algorithm, and after the workpiece coordinate system is established, a visual system is moved to the position above each programming hole site to shoot actual marks, and the workpiece coordinate system is calculated. According to the method, actual hole center coordinates are extracted through image recognition and compared with theoretical coordinates, and finally, deviation calculation and machining decision are carried out by integrating measuring head data, visual images, numerical control programs and tolerance parameters, so that through the automatic process integrating measurement, recognition, comparison and decision making, the machining precision is improved, and the machining precision is improved. The technical problems of manual errors and lack of verification before processing are effectively solved.

The invention discloses a bionic visual positioning system and method based on multispectral perception and a storage medium, and the system comprises a front-end multidimensional optical perception module, a mesopic vision preprocessing module and a rear-end fusion perception and target positioning module, the front-end multi-dimensional optical sensing module is used for performing optical imaging, spectral decomposition and polarization imaging; the mesopic vision preprocessing module is used for carrying out image preprocessing and optical flow estimation; and the rear-end fusion perception and target positioning module is used for feature fusion and target identification and positioning. According to the bionic visual positioning mechanism, through a bionic compound eye structure + multispectral + polarization + optical flow fusion perception strategy, the perception bottleneck of a traditional visual system in a high-dynamic and complex illumination environment is effectively solved, and the bionic visual positioning mechanism has the advantages of high robustness, low delay, wide field of view, multiple modes and the like; the method is especially suitable for application scenes such as automatic driving, unmanned aerial vehicles, robots and the like with extremely high requirements on real-time performance and reliability.

The invention discloses a non-consistency multi-focus unit modification system for laser slicing of transparent materials. The non-consistency multi-focus unit modification system comprises an ultra-short pulse laser, a light beam regulation and control system, a light beam shaping system, a focusing objective lens, a visual positioning system and a high-precision three-dimensional motion platform, a transparent material sample wafer to be sliced is placed on the high-precision three-dimensional motion platform; the ultrashort pulse laser provides picosecond laser beams for the whole system, the laser beams generated by the ultrashort pulse laser are transmitted to the light beam shaping system through the light beam regulation and control system to generate multiple inconsistent laser beams, and the multiple inconsistent laser beams are focused and shrunk through the focusing objective lens to obtain multiple inconsistent focuses. The focus is focused in a transparent material sample wafer on a high-precision three-dimensional motion platform for slicing; wherein the high-precision three-dimensional motion platform realizes the movement of a transparent material sample wafer in x, y and z directions; and the visual positioning system is used for identifying and positioning the transparent material sample. On the premise that the slicing quality is guaranteed, the processing efficiency is remarkably improved.

The invention relates to spraying equipment and a spraying method, in particular to intelligent spraying equipment and a spraying method, and the intelligent spraying equipment comprises the intelligent spraying equipment which is composed of a base, a multi-axis linkage mechanical arm, a precise spraying assembly, a visual positioning module, a rotary clamping mechanism and a central controller. Wherein the precise spraying assembly comprises a piezoelectric ceramic valve and an atomizing nozzle, and independent control over the micro flow and the atomizing angle can be achieved; the visual positioning module assists in precise positioning; the rotary clamping mechanism supports multi-degree-of-freedom adjustment; and the central controller realizes closed-loop spraying control through a track compensation algorithm. The invention further provides a spraying method, through accurate control over spraying parameters, track optimization and energy consumption management, the spraying precision, the film thickness uniformity and the coating utilization rate are remarkably improved, meanwhile, the overall energy consumption is reduced, and the high-quality spraying requirement of the complex workpiece surface is met.

The invention relates to the technical field of automation, in particular to a real-time multi-target visual positioning and mechanical arm cooperation system for intelligent material sorting, which comprises a visual perception and information acquisition module, a data processing and path planning module, a mechanical arm execution module and a man-machine interaction module. By combining the 2D camera and the 3D camera, 360-degree full-view-angle material sensing can be achieved, and the possibility of repeated grabbing errors of the mechanical arm can be reduced through the design of judging whether grabbing succeeds or not after the mechanical arm is controlled to execute the grabbing step every time.

The invention provides a GIS partial discharge positioning method based on deep learning and acoustoelectric combination, and relates to the field of data processing. The method comprises the following steps: firstly, acquiring an original ultrahigh frequency signal and an original ultrasonic signal of a target detection entity, constructing an acoustoelectric signal matrix, and performing short-time Fourier transform to obtain a complex spectrum tensor and a power spectrum tensor; and then, extracting a high-dimensional feature embedding tensor by using a double-branch deep learning network, and generating a time-frequency masking tensor and a source confidence tensor through a multi-source sub-ion network to complete source-by-source spectrum reconstruction. And a time difference of arrival matrix is calculated by using a phase transformation generalized cross-correlation method, a sensor space coordinate is combined to be input into the depth regression positioning sub-network, and a discharge source is output to predict three-dimensional coordinates and regression uncertainty. And finally, fusing the regression uncertainty and the source confidence tensor to construct a confidence interval, and generating a visual positioning result with an ellipsoid confidence boundary. By implementing the technical scheme provided by the invention, the accuracy of partial discharge positioning is improved.

The invention discloses a flight laser marking system and a flight laser marking method, and relates to the technical field of laser marking. The flying laser marking system comprises a feeding mechanism, a conveying mechanism, a marking visual positioning module, a laser marking machine, a visual detecting and positioning module, a defective product removing mechanism, a discharging mechanism, a sensing module and a main control and data processing system. The conveying mechanism comprises a product bearing belt and a synchronous belt. The sensing module comprises an encoder and a light reflection sensor. And the product supporting belt adopts a longitudinal groove array structure or a single / multiple transverse groove structure. The method comprises the steps of feeding control, visual positioning before marking, Y-axis coordinate calibration compensation, visual quality detection after marking, defective product removal, discharging control and the like. Continuous and full-automatic production is achieved, the production efficiency is greatly improved compared with static marking, meanwhile, the marking precision is guaranteed through the visual positioning and compensation technology, and online product quality control and defective product removal are achieved.