204 results about "Visual identification" patented technology

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 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 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 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 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 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 relates to the technical field of bridge construction monitoring, and discloses a bridge construction crack dynamic detection method and system based on AI visual identification. According to the method, pixel-level time domain and space domain feature separation and fusion are carried out by analyzing an obtained multi-frame continuous image sequence, a target structure feature map fused with spatio-temporal context information is generated, and a potential crack region is positioned according to the target structure feature map. And constructing a region evolution model by analyzing a pixel intensity multi-direction discrete change rule of the candidate region, and outputting quantitative description of a crack growth trend by using a crack growth prediction model. And a structural integrity dynamic score is generated through comprehensive growth trend and multi-scale texture analysis, and real-time detection and evaluation of the crack are achieved. According to the method, the accuracy and robustness of crack detection in a complex construction environment are improved, and the dynamic development trend of the crack can be effectively predicted.

The invention relates to the field of target feature recognition and unmanned aerial vehicle cluster formation, in particular to an unmanned aerial vehicle cluster using visual recognition beacons, which comprises a plurality of unmanned aerial vehicle sub-clusters. The unmanned aerial vehicle sub-cluster is composed of a central unmanned aerial vehicle and a plurality of auxiliary unmanned aerial vehicles. The auxiliary unmanned aerial vehicles carry visual identification beacons and fly within the visual range of the central unmanned aerial vehicle in a marshalling mode. The central unmanned aerial vehicle uses a visual system to actively detect and identify the visual identification beacons on the auxiliary unmanned aerial vehicles, obtains the position information of the auxiliary unmanned aerial vehicles, completes formation calculation according to task planning, generates a control instruction, and issues the control instruction to the auxiliary unmanned aerial vehicles, thereby achieving the efficient cooperation and agile formation of the cluster. The invention also comprises a method. According to the method, identity recognition, accurate pose calculation and motion state estimation of a large-scale cluster can be realized, rapid and high-agility formation transformation of the cluster is further supported, the method has the advantage of remarkable low detectability, expensive equipment does not need to be additionally arranged, and unification of high performance and low cost is effectively realized.

The invention discloses a robot visual identification decision control method based on deep learning, and relates to the technical field of deep learning, image data of each target object in a target area is collected, a data training sample and a data verification sample are established according to the collected image data, a conversion relation between coordinate systems of a binocular camera is constructed, and a target object is identified according to the conversion relation. A distortion coefficient, an internal parameter matrix and an external parameter matrix of the binocular camera are obtained, a convolutional neural network model is constructed, training is performed according to the data training sample and the data verification sample, a target object is identified through the trained convolutional neural network model, the horizontal distance between the target object and the robot is calculated, and clamping operation is executed. Feature extraction is carried out on image data through deep learning, the defect that a traditional method depends on manual setting of feature parameters is avoided, and the accuracy and adaptability of target object recognition are improved by importing the standard feature value of the target object and combining a comprehensive similarity calculation method.

The invention provides an intelligent cutting device for an underwater canal slope lining plate and a construction method. The device comprises an underwater self-adaptive walking platform, a three-dimensional visual recognition module, a cutting execution module, a buoyancy and pressure adjusting module and a remote control module. The intelligent cutting device for the underwater canal slope lining plate integrates three-dimensional laser scanning and algorithm recognition, the cutting range can be automatically recognized, manual judgment is not needed, the limitation of manual control of a simple robot is broken through, and the construction efficiency is improved; a triangular crawler belt of the underwater self-adaptive walking platform is matched with buoyancy adjustment of a buoyancy and pressure adjusting module, the underwater self-adaptive walking platform can be matched with a 1: 1.5-1: 3 variable-gradient channel slope and the water depth of 1-3 m, and the problem that a traditional robot slips and rolls over is solved. Closure pumping drainage is not needed, underwater cutting is directly carried out, the technical problem that water supply is affected by dry field repair is solved, and the pumping drainage cost is saved; no person enters the whole process, and safety accidents of divers are eliminated.

The utility model discloses a three-dimensional visual identification device for auxiliary material warehousing detection, which belongs to the technical field of auxiliary material warehousing identification and comprises a 3D camera, a visual support frame and a 2D camera. The visual supporting frame comprises a 3D camera support and a 2D camera support. The 3D camera is installed on the 3D camera support, and the 2D camera is installed on the 2D camera support. The 3D camera is used for carrying out high-angle shooting on the incoming materials, the two 2D cameras are used for carrying out auxiliary shooting on the incoming materials from the two sides, the types and the number of the incoming materials can be recognized more accurately and rapidly, the incoming materials at any time can be processed in time, dependence on personnel is small, and checking operation can be continuously carried out for a long time.

The utility model discloses a kind of visual identification's steel magnetic particle flaw detection mechanism, it is related to magnetic particle flaw detection technical field, the problem that the background technology proposed only to specific length workpiece can be carried out flaw detection, cannot be carried out all-around flaw detection to length longer cylindrical workpiece, low practicality, present following scheme, including magnetic particle flaw detection machine ontology, rack, operation platform and magnetizer, the operation platform top is equipped with two conveying mechanisms, the conveying mechanism includes the mounting seat on the outer wall of operation platform top by bolt connection, fixed block on the outer wall on the top of mounting seat by bolt connection, fixed plate on the outer wall on the top of mounting seat by bolt connection, conveying motor on the outer wall on the lower outer wall of one side of fixed plate by bolt connection and the conveying roller of rubber ring fixed on outer wall.Better practicality is improved by the utility model to the length longer cylindrical workpiece can be carried out continuous conveying, and all-around visual flaw detection detection is carried out to workpiece.

The invention provides a door access detection method and device based on a visual language model. The method comprises the following steps: acquiring a scene video, wherein the scene video comprises videos of access events of one or more room doors in a target scene; obtaining a visual identifier of the one or more room doors in the target scene, wherein the visual identifier indicates the position of each room door in the one or more room doors in the target scene; and inputting a prompt text and at least one of the following items into the visual language model to detect a door access event in the scene video: (i) an identification image comprising the visual identification and the scene video; and (ii) an identification video comprising the visual identification, wherein the identification video is obtained by identifying the scene video by using the visual identification.

The aerial work charging transfer system comprises at least one mooring unmanned aerial vehicle, a ground power supply unit and a ground control center, and an aerial charging interface device, a charging cable winding and unwinding device, a visual recognition unit and a flight control module are carried on the mooring unmanned aerial vehicle; the ground power supply unit is connected with a power storage device of the mooring unmanned aerial vehicle through a mooring cable, the ground control center is in communication connection with the mooring unmanned aerial vehicle, a winding roll, a driving motor and a sliding ring are arranged in the charging cable winding and unwinding device, and a charging cable connected between the mooring unmanned aerial vehicle and equipment to be charged is wound on the winding roll. The end of the charging cable is connected with the aerial charging interface device, and the visual recognition unit and the driving motor are electrically connected with the flight control module. The mooring unmanned aerial vehicle can quickly respond to the charging requirements of different positions and different heights, the shutdown time of equipment waiting for charging is greatly shortened, and potential safety hazards such as electric leakage, short circuit and stumbling caused by a long-distance and multi-connector temporary cable are avoided.

The application relates to the technical field of visual identification, in particular to a three-dimensional reconstruction device and method based on binocular camera and laser radar fusion under visual guidance, aiming to solve the problem of unstable three-dimensional reconstruction precision of binocular camera and laser radar fusion. The method comprises the following steps: acquiring two-dimensional images of left and right views of a field of view region; if there is a region of interest in the two-dimensional images, the pixel position of the region of interest is extracted, and based on preset calibration parameters and the pixel position, the azimuth angle and the pitch angle of the region of interest relative to the optical axis of the device are obtained; a scanning control signal is generated according to the azimuth angle and the pitch angle, and is output to a light beam deflection scanning unit; distance information of the region of interest is acquired, and a three-dimensional reconstruction model of the region of interest is obtained according to the distance information and the angle information. The three-dimensional reconstruction device and method based on binocular camera and laser radar fusion provided by the application improve the matching stability and reconstruction precision of complex regions such as weak texture, strong reflection and occlusion.

The present application relates to a kind of base station direct current comprehensive management and control device, belong to communication base station direct current power supply technical field.The device includes switching power supply, metering control module, direct current magnetic latching contactor, schottky diode, main control module, 4G communication module, visual identification module, base station standby battery, OLT standby battery, camera, infrared sensor, ultrasonic sensor, mains detection module, bluetooth module and the structure of FSU.The present application not only can realize the comprehensive management and control of base station direct current, can be carried out direct current management and control to single operator direct current load, also can realize the real-time monitoring of remote mountainous area and urban public security monitoring blind spot area communication base station, easy to popularize and apply.

The invention discloses an automobile underground receiving trough material level monitoring system and a control method thereof, the automobile underground receiving trough material level monitoring system comprises an underground receiving trough, an electronic belt scale, a dump truck, a truck scale, a camera, a display screen, a three-color lamp and an electric control system, the electronic belt scale is arranged below a blanking port of the underground receiving trough; the dump truck is used for feeding materials to the position above the underground material receiving groove. The truck scale is arranged below the unloading position of the dump truck; the camera is arranged above the unloading position of the dump truck; the display screen and the three-color lamp are arranged above one side of a cab at the unloading position of the dump truck; the system has the beneficial effects that the material level of the underground material receiving groove of the automobile is automatically monitored instead of manual work through automatic and visual identification technical means, the material level state of materials in the material groove can be continuously and completely reflected, production scheduling and automatic control over feeding and discharging of the material groove are facilitated, production continuity is guaranteed, and production benefits are increased.

The application discloses a kind of stem and leaf intelligent separation and high-precision measuring device and method, and it is related to plant phenotype analysis technical field.The device includes clamping module, weighing module, visual identification module, intelligent separation module, stem processing module, leaf processing module, chopping module, drying and weighing module and central control system.After clamping module fixes plant sample, weighing module obtains initial fresh weight;Visual identification module constructs stem and leaf point cloud model by three-dimensional scanning;Intelligent separation module plans cutting path based on model and executes separation;Stem processing module measures stem diameter and fresh stem weight;Leaf processing module flattens leaf and measures leaf area;After stem and leaf are chopped, send into drying and weighing module, and determine dry weight stable point by automatic weighing.The application integrates separation, measurement, drying function in one, realizes full-process automation, solves the problem of low efficiency, sample easy to damage, poor data consistency of traditional method, significantly improves the precision and efficiency of plant phenotype parameter acquisition.

The invention discloses an action indicating device for a fuse link and power grid equipment, and relates to the field of fuse structures, and the action indicating device for the fuse link comprises a main fuse body; the auxiliary melt is arranged in parallel with the main melt, the resistance of the auxiliary melt is far greater than that of the main melt, and the melting point of the auxiliary melt is far lower than that of the main melt; the linkage assembly is connected with the auxiliary melt; and the energy storage assembly is matched with the linkage assembly, and the linkage assembly is used for extruding the energy storage assembly when the auxiliary melt is not fused. According to the action indicating device for the fuse link, the main fuse body and the auxiliary fuse body are arranged in parallel, and after the main fuse body is fused due to overload or short circuit faults, circuit current is transferred to the parallel auxiliary fuse body branch, so that the auxiliary fuse body is fused rapidly, and the energy storage assembly releases all energy instantly; and the indicating piece is pushed to generate a clear visual identifier or a switching value signal.

The invention relates to the technical field of pathological sample management, and discloses a tumor pathological sample intelligent sorting and storage management system, which comprises the following modules and components: a sample information acquisition and right confirmation module, an AI intelligent sorting module, an intelligent storage module, a real-time state monitoring module and a full-chain tracing and auditing module. The sample information acquisition and right confirmation module comprises an RFID label printer, a high-cleaning code gun, a pathological information input terminal and a data right confirmation subunit; the AI intelligent sorting module comprises a six-axis sorting robot, a sorting conveyor belt, a visual identification camera and a sorting path planning algorithm unit; the intelligent storage module comprises a storage cabinet electric control door, a sample state monitoring subunit, a constant-temperature and constant-humidity storage cabinet and an RFID positioning unit; the real-time state monitoring module comprises a temperature and humidity sensor, a sound and light alarm device, an equipment state monitoring panel, a vibration sensor and an emergency intervention subunit; the problem that single path planning is mostly adopted in the sorting link is solved.

The invention discloses a navigation mark visual identification and collision early warning method and system based on deep learning, and relates to the technical field of intelligent shipping, and the method comprises the steps: collecting a navigation channel monitoring video stream in real time, and carrying out the preprocessing of a key image frame; constructing a navigation mark detection model based on an improved Officient Det network, and outputting the position and category of the navigation mark in the key image frame; the method comprises the following steps: constructing a ship track prediction model through ship historical track data, predicting a short-time track of a ship, calculating the relative position, speed and course angle of the ship and a navigation mark, and constructing a dynamic collision risk field; and fusing the visual detection result, the radar ranging data and the AIS information to obtain a ship collision risk, and triggering graded early warning based on a preset risk threshold. According to the method, the improved deep learning model and the real-time calculation framework are combined, the robustness of navigation mark identification is improved, an efficient collision early warning mechanism is established, and technical support is provided for intelligent shipping.

The utility model relates to the technical field of unmanned aerial vehicle inspection, in particular to an intelligent inspection device for a water taking line, which comprises a signal relay positioning auxiliary assembly and an inspection unmanned aerial vehicle assembly, the signal relay positioning auxiliary assembly is distributed along a water taking line inspection line of the inspection unmanned aerial vehicle assembly; the ground cable supplies power to the signal relay positioning auxiliary assembly. According to the utility model, the unmanned aerial vehicle monitoring module of the signal relay positioning auxiliary assembly is used to realize monitoring and radar positioning of the surrounding environment and the inspection unmanned aerial vehicle assembly, and four groups of multifunctional probe modules of the unmanned aerial vehicle monitoring module are used to realize day and night monitoring of the environment and a visual identification function of the inspection unmanned aerial vehicle assembly. Through cooperation with a built-in radar module of the unmanned aerial vehicle monitoring module, beyond-visual-range remote positioning tracking monitoring is carried out on the inspection unmanned aerial vehicle assembly, the work positioning precision of the inspection unmanned aerial vehicle assembly is improved, and a remote control platform of the inspection unmanned aerial vehicle assembly is assisted in completing accurate positioning of the flight route of the unmanned aerial vehicle.

The invention discloses an unmanned ship docking method and system based on air-ground collaborative visual identification. An unmanned aerial vehicle-mounted visual system collects a first position image of an unmanned ship and a dock cabin; carrying out target detection model detection, connecting an auxiliary line according to key points of a target obtained through detection, and calculating a relative distance and a course angle between the dock cabin door and the unmanned ship; whether the relative distance and the course angle reach a preset range or not is judged, the unmanned ship is controlled to move through the first propelling speed and the first course adjusting angle or the second propelling speed and the second course adjusting angle, and docking of the unmanned ship is repeatedly achieved. Through visual systems of the unmanned aerial vehicle and the unmanned ship, air-ground collaborative visual identification is realized. The images of the unmanned ship and the cabin door are captured in real time, the inclination angle and the relative position of the unmanned ship and the cabin door are accurately recognized in combination with a rotating target detection algorithm, and the safety and the efficiency of automatic docking of the unmanned ship are improved.

The invention discloses an airport bird identification and tracking method and system based on visual radar fusion, and the method comprises the steps: carrying out the three-dimensional positioning and morphological analysis of an airport airspace target through the cooperative work of a binocular stereo camera and a millimeter-wave radar; a background difference technology is adopted to extract candidate targets, and multi-dimensional descriptors are constructed through visual features such as morphological asymmetry and wing flapping frequency; performing distance and speed verification on a visual identification result by using micro-Doppler features of radar echoes, and determining a bird identification target through fusion error evaluation; an inter-frame image matching and shielding detection technology is implemented to generate a continuous trajectory sequence, and trajectory fluctuation parameters and periodic features are extracted to reflect flight dynamics; according to the method, identification inconsistency parameters are obtained through consistency checking of multi-frame identification results, bird target identification tracking results are output according to quality grading, and reliable information support is provided for airport airspace safety early warning and bird repelling decision making.

The invention discloses a blinding expelling method and system for a black flight unmanned aerial vehicle, a medium and a product, and relates to the technical field of unmanned aerial vehicles. The method comprises the following steps: inputting the model, the surface reflectivity, the initial surface temperature, the real-time distance and the environment illumination intensity of a target black-flight unmanned aerial vehicle into a preset decision model, and obtaining the optimal laser wavelength and the initial laser power matched with the target black-flight unmanned aerial vehicle for expelling the target black-flight unmanned aerial vehicle; dynamically focusing the laser beam according to the real-time distance, aligning a light spot of the laser beam to a photosensitive element area of the target black-flight unmanned aerial vehicle, and monitoring the actual output power of the laser and the surface temperature of the target black-flight unmanned aerial vehicle in real time; the driving current of the laser is adjusted through the PID controller; and when the visual identification signal detects that the target black flight unmanned aerial vehicle is in the out-of-control state for a preset time length, laser beam emission is stopped. By implementing the technical scheme, the expelling efficiency and safety are improved.

This application provides a fiberglass winding device, which includes a frame, a mounting base, a rotating sleeve, multiple ceramic wheel modules, and a mandrel. A metal component has multiple clearance holes arranged along the annular direction of the metal component. A metal induction switch is mounted on the support base and is used to sense the metal component rotating with the ceramic wheel to determine the rotational state of the ceramic wheel. Fiberglass fed by the multiple ceramic wheel modules is wound around the surface of the mandrel along the annular direction. When the metal component rotates, the metal induction switch continuously senses the metal component and outputs a sensing signal. When the metal induction switch is positioned relative to the clearance holes, it cannot sense the metal, causing the sensing signal to be interrupted, thus determining the interruption state of the fiberglass. This achieves autonomous identification of broken wires, avoiding reliance on human visual identification, reducing labor costs, and ensuring the effectiveness of the fiberglass winding device in identifying broken wires.

The invention relates to a backlight detection method and system, electronic equipment and a storage medium, and the method comprises the steps: obtaining a preset backlight execution parameter, and generating a backlight control instruction based on the backlight execution parameter; receiving a backlight control instruction, and performing differential control on the plurality of backlight areas divided by the backlight execution unit according to the backlight control instruction; and collecting a light-emitting effect picture of the backlight area, performing feature extraction and anomaly recognition on the light-emitting effect picture through a visual analysis algorithm, and outputting an anomaly detection result of the corresponding light-emitting element in each backlight area. By automatically controlling the on-off state of the backlight subareas and the brightness of the lamp beads and combining a visual identification algorithm to analyze the working state of the backlight light-emitting element in real time, efficient and accurate detection of the backlight light-emitting element is realized; the manual participation degree is effectively reduced, and the efficiency and accuracy of backlight detection are remarkably improved.

The invention belongs to the technical field of industrial detection and sorting technologies, and particularly relates to an aluminum profile surface defect online detection and sorting method which comprises the following steps: acquiring at least one framed picture when a target on a main conveying line is sorted for the first time; for each framed picture, processing the framed picture based on a picture recognition model to judge whether a defect exists on the surface of the target or not and determine the position of the defect; sorting the targets of which the defects are identified to a secondary sorting line; aiming at the position of the defect area, acquiring a three-dimensional point quantization parameter set of the defect; fine classification and cause tracing are carried out on the defects by using a multi-branch neural network; according to the method, accurate evaluation and flexible sorting of the surface defects can be realized by fusing two-dimensional visual identification, three-dimensional shape quantification and multi-stage intelligent decision, so that the differentiated quality requirements of customers are met to the maximum extent on the premise of ensuring the production feasibility, and the high-quality product rate and the resource utilization rate are improved.