332 results about "Visual identification" patented technology

The invention relates to the technical field of unmanned aerial vehicle obstacle avoidance, in particular to an AI-assisted unmanned aerial vehicle low-altitude flight obstacle avoidance method and system. The method comprises the following steps: carrying out real-time unmanned aerial vehicle low-altitude flight environment perception based on a multi-modal perception network, and carrying out static obstacle identification and obstacle radiation range analysis to obtain a plurality of static obstacle radiation paths; carrying out dangerous potential energy field modeling based on the plurality of static barrier radiation paths, carrying out dangerous environment distribution fitting, and constructing a three-dimensional dangerous potential energy map; performing multi-path flight rehearsal according to the three-dimensional danger potential energy map, performing optimal flight path evaluation, and extracting an optimal flight path; and performing unmanned aerial vehicle flight processing based on the optimal flight path, performing dynamic obstacle visual identification and operation path prediction, and constructing a plurality of obstacle movement prediction paths. According to the invention, the task execution efficiency and safety of the unmanned aerial vehicle are improved through rapid and accurate obstacle avoidance decision making of the unmanned aerial vehicle.

The invention relates to an intelligent visual identification technology-based cold solder joint microcrack intelligent discrimination method, which comprises the steps of collecting an initial RGB image of a to-be-detected welding spot, carrying out two-dimensional discrete cosine transform and inverse two-dimensional discrete cosine transform on the initial RGB image to obtain an enhanced image, and fusing the enhanced image with an R channel of the initial RGB image to obtain a fused image; forming a dual-channel feature map; calculating the phase consistency of the dual-channel feature map, and obtaining a suspected candidate region of the pseudo soldering microcrack through an adaptive threshold segmentation method; acquiring an RGB image sequence of a continuous time sequence of the welding spots, and performing anomaly detection to obtain an abnormal region set; and constructing a welding spot thermal diffusion model, and inputting the geometric parameters and the environmental parameters in the abnormal region set into the welding spot thermal diffusion model to obtain a final judgment result of the pseudo soldering microcracks. According to the method, through multi-dimensional feature fusion and continuous time sequence dynamic tracking, the detection precision of the pseudo soldering microcracks is remarkably improved, the false detection rate is reduced, and the final judgment result is more accurate.

The invention discloses a mobile phone silica gel shell appearance detection system based on an optical image sensor, and relates to the technical field of appearance defect detection and intelligent visual identification. An image acquisition module acquires a high-quality multi-channel image through multi-angle synchronous acquisition; the dynamic light field regulation and control module realizes self-adaptive regulation of a regional light source based on reflectivity perception and a gray gradient change rate; the image processing and defect identification module fuses edge features, texture changes and a deep learning model to realize pixel-level defect classification; the material identification and shielding module identifies a non-silica gel area through multi-dimensional spectral features and eliminates interference; a time sequence defect evolution analysis module models a defect evolution track; the trend prediction module is used for predicting the extension trend of potential cracks and fatigue areas; the control feedback module realizes intelligent judgment and response linked with the manufacturing system; the system can realize high-precision and high-robustness automatic defect detection and risk prediction, and is suitable for online quality monitoring of large-batch flexible products.

The embodiment of the invention provides a warehouse body intelligent unblocking method and system based on visual identification and air cannon linkage, and the method comprises the steps: collecting images in a warehouse in real time through a plurality of high-definition anti-explosion cameras disposed in the warehouse, carrying out the fusion processing of the collected images in the warehouse, and constructing a warehouse three-dimensional dynamic model; based on the warehouse three-dimensional dynamic model, key features are extracted by adopting a neural network; taking the extracted key features as input, analyzing associated parameters through a time sequence of a long-short-term memory network, establishing a congestion cause knowledge base, and labeling congestion types in a classified manner; constructing a congestion risk grading model based on the congestion cause library and a machine learning algorithm; and activating a corresponding air cannon based on the level corresponding to the three-level early warning and a visual positioning matching result, dynamically adjusting injection parameters, and executing unblocking operation according to an optimized time sequence. According to the embodiment of the invention, closed-loop intelligence from visual perception, reason modeling, artificial intelligence decision making to accurate execution is constructed.

The invention discloses a full-automatic intelligent port inspection robot system and method. The system mainly comprises an unmanned vehicle and an unmanned aerial vehicle which are autonomously coordinated. The unmanned vehicle platform comprises a vehicle body, an omni-directional driving wheel set arranged at the bottom of the vehicle body, and a multi-source sensing system, a 5G / Beidou dual-mode communication module, an emergency equipment cabin, an edge computing controller and an intelligent charging system which are arranged on the vehicle body, and real-time data interaction is achieved between the communication module and the unmanned aerial vehicle as well as between the communication module and the port central dispatching system. According to the invention, it is ensured that the robot system can complete core inspection tasks such as container three-dimensional scanning, container body state monitoring and storage yard path planning in an all-weather manner, automatic docking of the unmanned aerial vehicle and the unmanned vehicle with emergency equipment is realized, and a'sensing-decision-execution 'closed-loop system is formed; the functions of high-precision visual identification, multi-mode environment perception, dynamic obstacle avoidance planning, convenient replacement of unmanned aerial vehicle equipment, intelligent charging and the like are integrated, the port inspection efficiency can be remarkably improved, and the new-generation intelligent port equipment of the emergency response process can be optimized.

The utility model discloses the field of agricultural machinery, and particularly relates to a spraying machine tool and an all-terrain spraying vehicle, the spraying machine tool comprises a hollow shell, a visual identification device and a spraying device are arranged on the two sides of the shell in the width direction respectively, the visual identification device comprises a supporting rod, the supporting rod extends in the height direction of the shell, and the spraying device is arranged on the supporting rod. A camera is arranged at one end of the supporting rod, the other opposite end of the supporting rod is connected with the shell, the spraying device comprises a water pipe, the water pipe extends in the height direction of the shell, a signal control switch valve is arranged on the water pipe, a spraying pipe is connected to the top of the water pipe, and a plurality of nozzles are arranged on the spraying pipe at intervals. According to the utility model, the visual identification device and the spraying device are arranged, and the visual identification device is used for judging whether spraying is carried out or not, so that the pertinence is stronger during spraying, the investment of labor cost can be reduced, the physical and psychological health of staff is greatly maintained, and the spraying device is simple in structure and convenient to popularize.

The invention belongs to the technical field of visual identification, and particularly discloses a thick plate weld groove orthogonal section contour three-dimensional point cloud reconstruction method and system. Comprising the following steps: dividing a three-dimensional point cloud of a groove contour into a plurality of parts according to geometric characteristics of a weld groove; receiving designation of a target part, extracting a point cloud subset at the left junction of the target part, and further fitting to obtain an edge line; confirming a point closest to the three-dimensional point cloud of the current frame on the edge line; calculating a tangent vector of the edge line at the nearest point as a normal vector of an orthogonal section; calculating the distance between each point in the three-dimensional point cloud of the current frame and the normal vector of the orthogonal section, and screening out all points which do not exceed a preset threshold value; and calculating a projection point of each screened point on the orthogonal section to jointly form a reconstructed thick plate weld groove orthogonal section contour. According to the method, the coordinate system orthogonally perpendicular to the welding seam in real time is constructed, the three-dimensional point cloud is projected to the orthogonally perpendicular plane, the orthogonal section contour of the groove of the welding seam is obtained, and the measurement error caused by the pose of the sensor is eliminated.

The invention discloses a robot automatic grabbing path planning method based on visual identification, and relates to the technical field of intelligent grabbing. The method comprises the following steps: acquiring multi-view visual data of a target scene, and generating a scene three-dimensional compact reconstruction model through a multi-modal image fusion algorithm; performing target detection and feature extraction on the model, and screening an optimal capture point in combination with a visual attention mechanism; constructing a dynamic environment obstacle probability map, updating an obstacle state through time sequence visual tracking, and quantifying an interference weight; an initial grabbing path is planned based on an improved fast expansion random tree algorithm, path smoothness constraints and robot joint movement limit parameters are introduced, and path nodes are optimized through a Bezier curve; visual servo feedback and path deviation prediction are fused, path parameters are corrected in real time, and a continuous movement track is generated. The method effectively adapts to the dynamic environment, gives consideration to path safety, smoothness and mechanical arm motion characteristics, and remarkably improves the grabbing success rate and operation reliability.

The invention provides a self-moving device and a recharging method thereof, the self-moving device comprises a visual module, and the method comprises the following steps: obtaining a to-be-processed image containing a charging station through the visual module; performing image processing on the to-be-processed image to obtain a target image so as to obtain a contour line of the charging station; controlling the self-moving equipment to be aligned with the charging station by using the contour line; and after the self-moving device is aligned with the charging station, the self-moving device is controlled to move towards the charging station, so that the self-moving device is in butt joint with a charging head of the charging station, and a recharging function is realized. According to the method and the device, the self-moving equipment can realize automatic recharging without a boundary guide line or a visual identifier, so that the reliability and the success rate of the recharging process are greatly improved, in addition, the special requirements on an external environment and the dependence on additional facilities are reduced, and the maintenance cost and the complexity of the system are reduced.

The invention provides a driving anti-slip optimization control method based on vehicle-road cooperation road fusion estimation, and the method comprises the following steps: S1, fusing vehicle-mounted and roadside visual recognition information, and obtaining a vehicle-road cooperation road state recognition result through employing a weighted voting method in combination with road invariant hypothesis and historical information assistance judgment; s2, adaptively adjusting parameters of a dynamic estimator by using a pavement state recognition result, establishing a sample set based on a pavement fusion estimation result, introducing a fuzzy clustering algorithm of particle swarm optimization to realize self-evolution optimization of the structure and the parameters of the estimator, and outputting a fused attachment coefficient estimation result; and S3, designing a slip rate and wheel acceleration combined control strategy to deal with potential slip in advance, and outputting a control torque result of a slip rate controller. Compared with the prior art, the method integrates visual perception and dynamic information to realize accurate estimation of the pavement state, supports advanced intervention of the slip rate controller before slip, and improves the driving stability of the whole vehicle.

The invention discloses a highway hazardous chemical substance liquid leakage accident monitoring method and a highway hazardous chemical substance liquid leakage accident monitoring system based on computer vision. The method comprises the following steps: arranging and installing an identifier configured for hazardous chemical substance vehicle identification on a vehicle for loading and transporting hazardous chemical substances; the identifier comprises an optical identifier configured to be used for improving the visual identification degree of the dangerous goods vehicle and a digital identifier configured to realize encrypted storage and quick reading of vehicle information; on the basis of an existing road monitoring system, a lightweight identification model configured for detecting the states of an optical identifier and a digital identifier in real time is deployed and loaded in the existing road monitoring system; performing visual positioning on the road surface projection pattern through a camera, and calculating the real-time position of the vehicle by combining map coordinates so as to supplement the positioning blind area of the GPS in the tunnel mountain area; accident probability calculation is carried out based on the identification state to determine whether the vehicle is suspected to have an accident; and when the accident is judged to be suspected, triggering an accident response process. And the accident response efficiency is obviously improved.

The invention relates to the technical field of logistics site intelligent management, and provides a site digital twin management method based on IOT and AI visual identification technology, and the method comprises the steps: collecting multi-source heterogeneous data in a logistics site through a sensor, an RFID device, a GPS device and a camera which are disposed in the logistics site; creating a digital twinborn model corresponding to the logistics site; performing deep association analysis on the multi-source heterogeneous data to obtain potential association relationships and mode information among the multi-source heterogeneous data; according to the multi-source heterogeneous data and the corresponding potential association relationship and mode information, adjusting parameters and a structure of the digital twin model; and obtaining logistics prediction information, intelligent configuration and scheduling information and optimization suggestion information corresponding to the multi-source heterogeneous data, so as to complete digital twinborn management of the logistics site in a digital twinborn model according to the logistics prediction information, the intelligent configuration and scheduling information and the optimization suggestion information.

The invention discloses a filter tip modeling parameter visual identification and adjustment method, and relates to the technical field of filter tip production, a laser diameter measuring instrument and a linear array camera are used for synchronously collecting peripheral images of a filter tip, and high-precision and consistent size and gray scale double baselines are generated; a texture segmentation technology is adopted to identify a suspected recess area and generate a recess position vector, a confocal displacement sensor is utilized to carry out local depth walk-up to obtain depth distribution data, and a size compensation coefficient is generated and a size matrix is corrected through Gaussian process regression; statistical threshold analysis is conducted on the corrected size matrix, when parameters exceed the limit, the gap of a forming nozzle is adjusted, the size stability of the filter tip is ensured, the adjustment result and related data are written into a manufacturing knowledge base, follow-up batch optimization is supported, and the measurement reliability and the production line efficiency are improved; tiny recesses on the surface of the filter tip can be effectively identified and compensated, and the measurement accuracy and the assembly stability of modeling parameters are ensured.

The invention discloses a multi-size wafer pre-positioning method and system based on visual identification, and belongs to the technical field of wafer pre-positioning, and the method specifically comprises the steps: collecting a to-be-positioned wafer image of a stage region, employing a visual feature extraction model to recognize the outer edge features of a to-be-positioned wafer, constructing an outer edge candidate curve, and carrying out the recognition of the outer edge candidate curve; performing multi-circle fitting on the outer edge candidate curve, determining a size parameter of a to-be-positioned wafer by adopting a hierarchical confidence rule, generating an eccentric correction path of the minimum rotation amount based on the size parameter of the to-be-positioned wafer, predicting the offset of the wafer relative to the center of the carrying table, and establishing a rotation compensation sequence; performing segmented rotation and micro-displacement adjustment on the carrying table according to the eccentric correction path, and performing pre-positioning on the wafer to be positioned under the condition of multiple sizes; the method does not need to depend on a template with a fixed size or a mechanical positioning structure, can automatically analyze the outer edge features of wafers with different sizes, and completes eccentric prediction and rotation compensation based on visual feedback.

The invention relates to a girth welding device and method for aviation pipeline parts, and belongs to the technical field of aviation manufacturing and welding, and the device comprises a supporting platform, a positioning assembly, a clamp assembly, a welding assembly and a control assembly. The positioning assembly determines the welding seam position through visual identification and laser alignment; the clamp assembly adopts a flexible clamping module and a clamp body to fix a to-be-welded pipeline; the welding assembly completes welding operation through a sliding rail and a welding gun module. The control assembly achieves accurate control over the motion trail and the output power of the welding gun module through the trail control module and the heat influence control module. Clamping can be adjusted in a self-adaptive mode according to the inner diameter of a to-be-welded pipeline, and deformation is avoided; the power of the welding gun module is adjusted through temperature monitoring, the heat influence is reduced, the welding quality stability is improved, the welding effect is further optimized through the circulating cooling system, and the requirements of aviation pipeline parts for high-precision and high-quality welding are met.

The invention discloses a plate positioning method and system based on visual identification, a robot feeding method and system and a computer program product. The plate positioning method comprises the following steps: acquiring a plate image containing a plurality of plates arranged side by side, wherein a label identification code is pasted on the surface of each plate according to a preset pasting direction; determining the number and the arrangement positions of the plates according to the number and the arrangement positions of the label identification codes in the plate images; scanning the label identification code on each plate to obtain the size information of the plate; according to the placement direction of the label identification code, determining the placement direction of the plate corresponding to the label identification code; performing overall contour recognition on the plate image to obtain an overall contour; according to the number of the plates, the arrangement positions of the plates, the size information of the plates and the placement directions of the plates, the overall contour is subjected to plate division, and contour position information of the plates is obtained. According to the plate positioning method, even if a plurality of pure black or pure white plates are placed side by side, the contour position information of each plate can be accurately identified and obtained.

The invention discloses a multi-band transient electromagnetic self-adaptive scanning dam leakage channel three-dimensional imaging method. The method comprises the following steps: arranging a plurality of detection nodes on the surface of a dam; the central control unit sends a scanning instruction to the detection node; the detection node scans the dam and uploads the data to the central control unit; and after the data is uploaded, the central control unit calculates an underground three-dimensional resistivity data volume through an inversion algorithm and performs rendering imaging. According to the invention, through combination of multi-band detection and intelligent adaptive scanning, efficient and accurate dam leakage channel three-dimensional visual identification is realized, and the defects of low efficiency, single detection dimension and rigid scanning strategy of a traditional method are effectively overcome.

The invention relates to the technical field of industrial automation, discloses an intelligent welding method and system for tube panel module production, and aims at solving the problems that manual welding is high in dependence, uneven in quality, low in efficiency, high in cost and insufficient in monitoring optimization capacity. The method comprises the steps that a workpiece connector is recognized through three-dimensional vision, and a path is planned; carrying out real-time monitoring, data fusion and feature extraction by a multi-modal sensor; deep learning is performed to predict quality, and reinforcement learning is performed to optimize parameters; and finally, post-welding evaluation and tracing are realized. The system comprises a main control unit, a robot welding arm, a welding execution unit and a positioning clamping unit. A multi-mode intelligent sensing module; and an intelligent processing and auxiliary control unit. By the adoption of the technical scheme, full automation, high efficiency, high precision, high consistency, quality real-time monitoring and intelligent optimization of key welding of the tube panel module can be achieved.

The invention discloses a famous high-quality tea target detection method based on an improved RT-DETR model, and relates to the technical field of visual recognition. Comprising the following steps: obtaining tea photos under different conditions, and preprocessing the tea photos to obtain a training set; based on a feature enhancement module, an improved multi-scale feature pyramid network module, a feature fusion module, a partial convolution and CGLU module, an OmniKernel module and a frequency domain-space domain feature enhancement module, improving the RT-DETR model and establishing an improved RT-DETR model, and training the improved RT-DETR model by using a training set; the trained improved RT-DETR model can be used for identifying the tea leaves. According to the invention, through multi-module coordinated optimization, the model keeps light weight, and MAP50 reaches 0.87, which is improved by 3.3% compared with a reference model. Especially under a complex background, the omission ratio is reduced by about 40% in tea target identification.

The invention belongs to the technical field of visual identification, and particularly discloses a thick plate saddle-shaped weld groove feature extraction and size measurement method and system. Comprising the steps that based on multi-scale geometric feature mutation detection, a current frame weld groove three-dimensional point cloud is segmented into a branch pipe face, a groove side wall, a backing weld surface and a main pipe face, and four area three-dimensional point clouds are obtained; according to the three-dimensional point cloud of each area, determining an intersection point of a branch pipe surface and a groove side wall, an intersection point of the groove side wall and a backing welding surface, an intersection point of the backing welding surface and a main pipe surface, an intersection point of the branch pipe surface and the main pipe surface, and a virtual intersection point of the groove side wall and an extension line of the main pipe surface as weld groove feature points; and constructing vectors between the feature points, and calculating the geometric dimension of the weld groove through vector operation. The method is based on multi-scale geometric feature mutation detection and adaptive point cloud segmentation, a fitting curve slope threshold value does not need to be preset, and the method can be adaptive to different included angles between the branch pipes and the main pipe in the pipe tubular joint.

The invention discloses a slope verification method based on unmanned aerial vehicle visual identification and related equipment, and relates to the technical field of image processing, and the method comprises the steps: collecting data through unmanned aerial vehicle multi-source equipment, and carrying out the fusion of the data to generate a multi-scale fusion image as a unified analysis substrate; secondly, extracting region features from the region features to carry out intelligent partitioning, identifying key element features, mapping the key element features to each partition, and establishing a region-element association model; thirdly, time sequence dimensions are introduced, current partition data are compared with historical slope structure data, an element reduction analysis method is innovatively adopted, mechanical parameters are dynamically corrected according to element development and evolution, and therefore the slope change amount and the structure deformation amount reflecting the real damage state are calculated; and finally, generating a slope thermodynamic diagram based on the quantitative indexes, and realizing risk visual output.

The invention discloses a first-level road non-ETC vehicle toll collection method and system based on multi-source data fusion, and the method comprises the steps: scanning a vehicle in a radar sensing recognition region in real time through a radar, so as to obtain radar scanning information; generating a vehicle tracking record based on the radar scanning information, wherein the vehicle tracking record comprises the target vehicle and first vehicle information corresponding to the target vehicle; in response to the target vehicle entering the visual identification area, an image acquisition instruction is sent to image identification equipment by using a radar, and the image identification equipment comprises license plate identification equipment and vehicle body identification equipment; acquiring second vehicle information acquired by the image recognition equipment; fusing the first vehicle information and the second vehicle information to obtain vehicle fusion information; and determining identification information of the target vehicle based on the vehicle fusion information, and determining a charging amount of the target vehicle according to the identification information. According to the invention, through fusion processing of the multi-source data, the accuracy of vehicle identification is effectively improved, and the conditions of misidentification and missed identification are reduced.

The invention relates to the technical field of deformation detection, in particular to a steel bar truss floor support plate deformation monitoring method and system based on visual identification, and the method comprises the following steps: installing an industrial camera at the top of a steel bar truss floor support plate to shoot a reference image, and installing an attitude angle sensor on a support to collect a reference attitude angle; when deformation is detected, a current image and a synchronous posture are taken, posture deviation is calculated, node pixel coordinates are corrected, offset is obtained by comparing reference coordinates, first ten nodes of offset are collected and screened, and the plane offset state of the floor support plate is output. An attitude angle sensor is arranged on a camera support to collect an initial pitch angle and a roll angle to establish a monitoring reference, a current image and a camera attitude are synchronously taken during deformation detection, attitude deviation correction node pixel coordinates are calculated, camera attitude systematic errors are offset, misjudgment of deformation is avoided, reference stability is guaranteed, and data accuracy is improved. And support is provided for deformation judgment of the floor support plate.

The invention discloses a tippler dynamic unhooking robot control method based on visual identification, and belongs to the field of industrial robots, and the method comprises the steps: deploying a visual camera array in a tippler operation area, collecting tippler-carriage image data, and carrying out the dynamic target identification of the tippler-carriage image data; constructing a compartment claw instance segmentation network to perform segmentation positioning on the dynamic instance target set; acquiring claw relative motion state data to predict claw motion, compensating a claw motion deviation prediction parameter to a vehicle claw connection position, and determining an unhooking operation position parameter; and performing smooth unhooking control on an execution mechanism of the dynamic unhooking robot of the car dumper through the fuzzy logic controller based on the unhooking operation position parameters. The technical problems that in the prior art, unhooking of a car dumper depends on manual assistance or a fixed program, the adaptability to dynamic conditions such as claw vibration and image interference is poor, and unhooking stability and accuracy are caused are solved.

The embodiment of the invention discloses a region boundary early warning method and system based on unmanned aerial vehicle visual identification, and the method comprises the steps: obtaining a sample cruise video, obtaining a sample video frame based on the sample cruise video, inputting the sample video frame into a detection model for feature extraction, and obtaining a backbone output feature; carrying out multi-path convolution processing on the backbone output features to obtain enhanced features, carrying out cross attention based on the enhanced features to obtain attention fusion features, carrying out object prediction based on the attention fusion features to obtain prediction object parameters of sample objects, and training a detection model according to differences between label object parameters and the prediction object parameters to obtain a detection result; the to-be-detected cruise video and the region boundary parameters are obtained, the to-be-detected cruise video is input into the trained detection model for real-time object detection, the target object parameters of the target dynamic object are obtained, the early warning level is determined based on the target object parameters and the region boundary parameters, and the accuracy of region boundary early warning can be improved.

The invention relates to the technical field of intelligent inspection and cooperative control, and particularly discloses a robot dog and unmanned aerial vehicle cooperative inspection system and method based on an artificial bee colony algorithm, and the system comprises an unmanned aerial vehicle, a robot dog and a cooperative decision system. The robot dog comprises a navigation module, an environment sensing module, a first communication module and a charging platform; the unmanned aerial vehicle comprises a flight control module, a visual identification module, an image acquisition module, an energy management module and a second communication module; the collaborative decision module is used for operating an artificial bee colony algorithm to realize intelligent collaboration of the robot dog and the unmanned aerial vehicle; according to the invention, through dynamic conversion of roles, active exploration of an unknown area and autonomous precise charging on the mobile platform, all-weather, self-organization and full coverage of the inspection process are realized, and the inspection efficiency and the intelligent level in a complex environment are significantly improved.

The invention discloses an intelligent inspection warning device for an automatic loading and unloading port wharf, and relates to the technical field of warning devices. A solar power generation panel module is fixedly installed in the middle of the outer wall of the identification conical cylinder and used for improving the cruising ability of the device. A yagi antenna is fixedly mounted at the top end of the identification conical cylinder; the outer wall of the metal keel inner cylinder is fixedly sleeved with a thermal insulation protection sleeve, and the thermal insulation protection sleeve is fixedly installed in the identification conical cylinder. According to the invention, the visual identification system, the microwave RFID system, the positioning module and the cloud end are integrated for cooperation, the processor processes environment information and cloud instructions by using a three-dimensional reconstruction algorithm, an optimal path can be autonomously planned, and centimeter-level movement precision is realized by means of ground marking identification; the system can automatically move to a designated position according to a cloud instruction, quickly responds to wharf area changes, such as a newly added operation area and a temporary blocking area, does not need manual carrying, and greatly improves the laying efficiency and flexibility.

The invention relates to a hydraulic support face guard state monitoring and early warning method and device based on three-dimensional modeling, and the method comprises the steps: carrying out the multi-conversion mapping among a world coordinate system, a camera coordinate system, an image coordinate system and a pixel coordinate system based on an original working condition image of a fully mechanized coal mining face in combination with a face guard three-dimensional model and folding / unfolding safety line information; generating a working condition image fused with the three-dimensional projection; a deep learning instance segmentation algorithm is adopted to extract the contour of the face guard, and a to-be-detected area is dynamically divided; the folding state of the face guard in the advancing direction of the coal mining machine and the unfolding state of the rear face guard are monitored by comparing the overlapping area of the contour of the face guard and the projection area of the safety line; and frame numbers and early warning information are automatically marked for the abnormal state face guard plates, and a visual early warning image is generated. According to the invention, through a double verification mechanism of three-dimensional space projection and two-dimensional visual identification, full-period state monitoring of the face guard is realized, hardware dependence is reduced, three-dimensional space characterization is enriched, and high monitoring accuracy is maintained in an underground complex environment.

The invention discloses an aluminum foil surface defect detection system based on visual identification, and relates to the technical field of aluminum foil detection, and the system comprises a data acquisition module which is used for obtaining image data and environment data in a detection area; the data processing module is used for processing the obtained image data to obtain a region fusion layer; the image layer optimization module is used for carrying out image layer boundary optimization on the obtained region fusion image layer and splitting the region fusion image layer after boundary optimization to obtain an aluminum foil feature image layer; the defect identification module is used for carrying out surface defect analysis on the aluminum foil characteristic pattern layer and outputting a surface defect analysis result; according to the method, environment data including various standards are set, dynamic matching and layer fusion are performed during detection, and a self-adaptive image analysis reference is constructed, so that imaging quality fluctuation caused by environment illumination change is reduced, and the detection accuracy and reliability in a changeable industrial field are remarkably improved.

The invention discloses an unmanned aerial vehicle accurate landing method based on multi-mode fusion perception, and relates to the technical field of unmanned aerial vehicle landing control. According to the method, an unmanned aerial vehicle needs to be guided to descend to a preset height through a first positioning module, then the unmanned aerial vehicle is switched to an airborne visual recognition module to recognize a ground visual target, if recognition succeeds, a multi-mode fusion model based on scale and space alignment is adopted to accurately calculate the offset, and the offset is calculated; and finally, a closed-loop control method based on visual servo is adopted, a speed instruction is generated and corrected in real time according to the calculated offset until precise landing at the target position is achieved, and therefore the centimeter-level landing precision, the whole-process autonomy and the high-safety landing process are achieved, and remarkable technical advantages and commercial application value are achieved.