155 results about "Visual control" patented technology

The invention relates to the technical field of visual control, in particular to a picture defect test equipment control system and method, and the system comprises a difference tensor construction module, a space-time constraint construction module, a strain response judgment module, a coupling focusing mapping module and a boundary stability evaluation module. According to the method, the multi-dimensional difference incidence relation is constructed by extracting the pixel gray difference value, the brightness gradient and the chrominance phase angle, and fine brightness fluctuation and color shift in a picture can be captured. The superposition of the time-space change rate enables the dynamic performance of image defects to be distinguished, and continuous and transient anomalies can be effectively recognized. And in combination with tensor strain ratio and difference amplitude analysis, an abnormal region can be adaptively recognized, and neglect of a traditional method on a complex scene is avoided. The weighted aggregation of the local energy response improves the stability of the defect boundary, ensures the accurate positioning and judgment consistency of the image defect, and significantly improves the reliability and accuracy of the detection result especially in dynamic change and complex illumination environments.

The invention provides a binocular vision teleoperation robot system, a robot and a teleoperation method and device, and relates to the technical field of computers. The method comprises the following steps: receiving a binocular video stream obtained by carrying out image acquisition on an actual scene by a binocular vision robot, and rendering the binocular video stream to obtain a three-dimensional scene; displaying the three-dimensional scene to an operator of the binocular vision robot to capture motion data of the operator observing the three-dimensional scene; determining a target area from the actual scene according to the action data, and obtaining a visual control instruction based on the target area; the vision control instruction is transmitted to the binocular vision robot for teleoperation; wherein the visual control instruction is used for adjusting imaging parameters of the binocular vision robot for image acquisition of an actual scene, and the visual direction of the binocular vision robot can be rapidly and naturally kept consistent with the attention point of the operator.

A system and method to generate a target image from a reference image are provided. The system may receive, via a LDM, a reference image and a text prompt. The system may extract, via a trained vision encoder in the LDM, a vision control signal from an object in the reference image. The vision control signal indicates an identity of the object. The system may extract, via trained text encoders in the LDM, text control signals associated with the text prompt. The system may generate, via cross attention summation of an output of a vision cross attention unit(s) associated with the vision control signal and an output of text cross attention units associated with the text control signals, spatial features indicative of the reference image and the text prompt. The system may output, via a decoder in communication with the LDM, a target image based on the generated spatial features.

The invention discloses a mechanical arm hand-eye calibration method based on a standard model and mark points, and belongs to the technical field of visual control, and the method comprises the following steps: based on a URDF standard model, determining a plurality of key points of a mechanical arm and local three-dimensional coordinates of the key points in each connecting rod coordinate system; for each moving image, identifying two-dimensional coordinates containing key points; determining global three-dimensional coordinates of the key points in a base coordinate system; calculating candidate calibration parameters according to the two-dimensional coordinates and the global three-dimensional coordinates of the key points, and fitting the candidate calibration parameters of all the moving images to obtain initial hand-eye calibration parameters; analyzing the actually measured poses of the mark points from the detection image; calculating theoretical poses of the mark points; and generating the hand-eye calibration parameters according to the actually measured poses and the theoretical poses of the mark points and the initial hand-eye calibration parameters. According to the invention, the problem that the hand-eye calibration of the robotic arm is not accurate enough in the prior art can be solved.

The invention discloses a mine environment data LED display method based on edge calculation, and the method comprises the steps: 1, collecting multi-source environment data, and carrying out the preprocessing and index block coding; step 2, a gating fusion feature tensor is generated through an improved EdgeViT network; 3, executing risk reasoning and evaluation through a risk perception module, and outputting a risk evaluation result; 4, when a communication abnormal state is detected, risk prediction reasoning is executed based on the historical cache; 5, if the collaborative model is deployed, fusing the risk assessment results of the improved EdgeViT model and the collaborative model; sixthly, a visual control instruction is generated through an LED display control module, and risk information is dynamically displayed; and step 7, collecting feedback information, and performing incremental updating on the EdgeViT network and the risk sensing module based on the feedback information. According to the invention, the real-time performance of mine environment monitoring, the reliability of risk early warning and the intelligent response level of LED display are improved.

The invention discloses a visual monitoring system for the repairing effect of a cable buffer layer repairing liquid based on multi-sensor information fusion. The monitoring system comprises a repairing liquid injection control module, a multi-sensor monitoring module, a data fusion analysis module and a visual terminal module. The device has the beneficial effects that a synergistic effect is formed through technical improvement of the four modules, the diffusion range and penetration depth of the repair liquid can be accurately captured in real time, waste and liquid accumulation caused by excessive injection of the repair liquid are avoided, and the blind area problem of a traditional repair technology is solved; meanwhile, a parameter database and a self-adaptive fusion algorithm are universalized, so that the adaptability of the technology to different scenes is greatly improved, and the limitation that the traditional technology is poor in universality is broken through; and through real-time dynamic map display and historical data tracing functions, alarm and injection control are linked to form a closed loop, so that visual control of the repair effect is realized, and the safety and controllability of the repair process are further guaranteed.

The invention relates to a space-time alignment and semantic association modeling method for cross-platform geographic information data, and belongs to the field of big data and natural language processing. The method comprises the following steps: step 1, carrying out space-time uncertainty modeling and probabilistic representation, and quantifying time and space uncertainty of multi-source data; 2, combining an alignment decision mechanism with adaptive learning to complete multi-evidence intelligent flexible alignment fusion; step 3, carrying out deep semantic analysis and real-time association discovery based on multi-modal semantic understanding and association mining; and 4, dynamically updating and reasoning the knowledge graph, and performing real-time analysis and real-time reasoning. According to the method, for a sudden search task, a new data source can be rapidly integrated, the knowledge graph is enriched, global situation visual control is achieved, multilevel reasoning based on the knowledge graph is achieved, efficient and accurate geographic information data intelligence is supported, and the task response efficiency is improved.

The invention discloses a method and device for generating a first-screen video background based on page elements, and the method comprises the steps: obtaining a multi-mode interface feature of a first screen of a target website, and receiving a natural language instruction inputted by a user through an interaction dialog box; analyzing the visual intention, style preference and dynamic demand in the natural language instruction to generate a structured visual control description; extracting layout, color and semantic entities from the multi-modal interface features through a cross-modal attention model to form and generate condition vectors; and inputting the generated condition vector into a condition decoupling video diffusion model, decoupling the generated condition vector into a control subspace of content, style and motion, and synthesizing a video background sequence according to the control subspace and the structured visual control description so as to synthesize a dynamic background in the first screen of the target website. According to the method, the dynamic video background is generated under the condition of the webpage interface and the guidance of the user language, so that the defect of rough control of a general text video model is avoided.

This invention relates to the field of cable monitoring technology, specifically disclosing a cable fire monitoring system and method based on distributed optical fiber sensing. The distributed optical fiber sensing unit is attached to the cable or its surroundings, simultaneously collecting temperature, vibration, and smoke signals and converting them into optical signals. After signal demodulation processing, an intelligent early warning and positioning unit, combined with algorithms, diagnoses the status and accurately locates the cable. A coordinated response unit automatically executes alarm, fire extinguishing, and power-off operations, while a remote monitoring unit enables multi-terminal visual control. This invention overcomes the limitations of single-parameter monitoring, offering high monitoring accuracy and fast response, effectively reducing false alarms and missed alarms. It is adaptable to various complex environments, possesses advantages such as high stability, low maintenance costs, and no need for on-site supervision, and can be widely applied to various cable fire monitoring scenarios to achieve early detection and early handling of potential hazards.

The invention discloses a mechanical arm high-precision movement control method based on double-path observation and online optimization. The movement precision of a mechanical arm is improved. The method comprises the following steps: step 1, acquiring an initial value of each joint angle of the mechanical arm; 2, an initial first conversion matrix of the tail end of the mechanical arm relative to a mechanical arm base is obtained; optimizing the initial first conversion matrix to obtain an optimized first conversion matrix; 3, an initial second conversion matrix of the tail end binocular camera relative to the tail end of the mechanical arm is obtained; optimizing the initial second conversion matrix to obtain an optimized second conversion matrix; 4, the optimized first conversion matrix and the optimized second conversion matrix are applied to a visual control system of the mechanical arm, a first coordinate value obtained by a tail end binocular camera and based on a tail end binocular camera coordinate system is converted into a mechanical arm base coordinate system, and a second coordinate value is obtained; and the tail end of the mechanical arm is controlled to move to the second coordinate value position.

This application provides a visual control method, system, storage medium, and program product for self-service baggage check-in equipment based on real-time baggage monitoring data, relating to the field of intelligent visual control technology. The method includes: acquiring baggage monitoring data in real time, including trigger signals from multiple photoelectric sensors and weight data output from a weighing platform; determining the speed evolution sequence of baggage based on the trigger signals and performing trend analysis; performing time-series coupling analysis of the trend analysis results and weight data to determine whether the baggage has entered a stabilization transition phase; predicting the target time when the baggage reaches a stationary state during the stabilization transition phase and sending a pre-configured command to the visual system in advance; and confirming that the baggage has reached a stationary state by verifying the fluctuation amplitude of the weight data after the target time arrives, and controlling the visual system to perform image acquisition operations. This solves the technical problem of poor visual recognition reliability of visual systems during the dynamic flow of baggage in related technologies.

The application relates to an indoor intelligent fire extinguishing robot and a chassis control method, which comprise a running system, a power supply system, a fire extinguishing system, a visual control system and a central controller; the running system is used for advancing, retreating and differential steering of the fire extinguishing robot; the power supply system is used for supplying power to the running system, the fire extinguishing system and the visual control system; the fire extinguishing system is used for extinguishing a fire source area; the visual control system is used for judging whether an obstacle can be normally passed when the robot runs, and if the obstacle cannot be normally passed, the local path needs to be re-planned to run to the fire disaster area, and the visual control system is used for identifying a fire source so that a fire extinguisher can be accurately sprayed to the fire source part; and the central controller is used for controlling the running system, the power supply system, the fire extinguishing system and the visual control system. The application has the advantages that rear wheel active steering and front wheel steering as an auxiliary are adopted, the differential steering has a small steering radius, and good steering stability is achieved.

The invention discloses a full-automatic color mixer, relates to the technical field of color mixing, and aims to solve the problems that traditional manual color mixing is insufficient in precision, and an industrial color mixer is large in size and high in cost. The visual control module comprises an integrated touch panel and is provided with an RGB scroll bar and an RGB conversion user library; the color conversion calculation module normalizes RGB, then calculates K / C / M / Y values and converts the K / C / M / Y values into percentage; a motor control panel calculates the volume and weight of each pigment; a power driving module four-color water pump is connected with a silica gel hose and a pigment barrel; after a user selects colors, the system automatically completes color conversion, pigment batching and mixing, and the color matching precision is delta Elt; 3; the shell is a PETG 3D printing split type shell, is small in size and is suitable for families and small working rooms.

The present application relates to the technical field of energy consumption control, and particularly relates to an energy consumption prediction and energy saving control system. The technical scheme comprises a data acquisition and communication module, an energy consumption prediction and optimization module, an intelligent control execution module, an energy management and visualization module, and a system coordination mechanism. The data acquisition and communication module is used for collecting operation parameters, environmental parameters and equipment states of building mechanical and electrical equipment in real time, and realizing data aggregation and transmission through a multi-protocol bus. The present application realizes the cooperative optimization of building energy consumption through dynamic prediction and intelligent control, responds to environmental changes in real time, and accurately coordinates the operation of multiple devices, thereby breaking through the limitations of traditional model hysteresis and coupled interference. In combination with multi-source data fusion diagnosis, hidden dangers are actively identified, operation and maintenance costs are reduced, standardized integrated design simplifies construction management, intuitive visual control is provided, and a full-cycle closed loop of energy efficiency improvement and intelligent operation and maintenance is realized.

The patent belongs to the technical field of battery pack disassembly, and relates to a battery pack disassembly method. An outer shell disassembly visual camera is used to collect point cloud data of a battery pack outer shell, a visual controller is used to obtain outer shell cutting track point data, and laser cutting is used, which is suitable for cutting of outer shells of different specifications of battery packs. A first visual camera is used to collect a depth map of an electrode connecting piece, a visual controller is used to process and obtain base point coordinates of electrode centers corresponding to each electrode connecting piece, and laser cutting is used. A second visual camera is used to obtain unloading coordinates of the cut electrode connecting piece, and the electrode connecting piece can be automatically unloaded. Power conveying lines are used between the stations to transfer materials, realize online processing, have high efficiency, high flexibility and good safety. The patent also provides a battery pack disassembly line.

The utility model relates to the technical field of video acquisition, and one embodiment of the utility model provides a training equipment indicating lamp state acquisition device which comprises a rectangular frame and an arc-shaped frame, the arc-shaped frame is fixed at the front end of the rectangular frame, an installation fixing assembly is arranged on the rectangular frame, a front arm is arranged outside the arc-shaped frame, and a rear arm is arranged outside the arc-shaped frame. The retractable assembly is arranged between the front arm and the arc-shaped frame, the telescopic collecting assembly is arranged on the front arm, the mounting and fixing assembly comprises a driving screw rod, the driving screw rod is connected into the rectangular frame in a threaded fit mode, one end of the driving screw rod is rotationally connected with a clamping plate in a sleeved mode, a pair of round rods is arranged on the side surface of the clamping plate, and the round rods are movably connected into the rectangular frame in a sleeved mode. By means of the technical scheme, the technical problems that in the prior art, a visual control system is used for shooting, recognizing and outputting, and the method has the defects that the system is complex, cost is high, and the method cannot be used on the basis of original practical training equipment are solved.

The invention belongs to the technical field of submarine underwater sealing, and particularly relates to a submarine through-board sealing and intelligent monitoring alarm device and a sealing pressure threshold setting method, and the device comprises a cup-shaped pipe joint (1), a through-board cable (7), a sealing filling assembly and a locking part used for pressing the sealing filling assembly; the locking component is a cable locking pressure intelligent measuring module (8); the cable locking pressure intelligent measurement module (8) comprises a sensor main body (81), a strain sensing unit and a signal processing and display unit (83). According to the method, efficient and reliable multi-layer high-pressure-resistant sealing can be provided, the locking pressure of a sealing interface can be accurately measured and displayed in real time, remote monitoring is supported, an alarm is automatically given when the pressure is abnormally reduced, and therefore intelligent and visual management and control over the through-board sealing state are achieved.

The invention discloses a deep learning-based target grabbing method of a robot, and relates to the technical field of robot visual control and industrial automation. Comprising the following steps: S1, calling an RGB camera, a depth camera and a polarization camera of a robot, synchronously collecting multi-modal visual data of a target object, and executing space-time synchronous calibration on the multi-modal visual data to form a space-time aligned multi-modal image data set; s2, performing optical interference preprocessing on the multi-modal image data set; according to the method, the RGB camera, the depth camera and the polarization camera are called to synchronously collect the multi-modal visual data and perform time-space synchronous calibration, so that the time-space alignment accuracy of the data is ensured; light reflection pseudo textures and depth noisy points are eliminated through optical interference preprocessing, and interference source influences are removed; a double-branch convolutional neural network is utilized, an initial morphological feature map is generated through a geometric feature extraction branch, and false texture interference is inhibited by combining attention weight map correction features generated by an optical feature correction branch.

The invention provides a nuclear power station pipeline cleaning robot. The nuclear power station pipeline cleaning robot comprises a main body, a walking mechanism, a cleaning mechanism, a collecting mechanism and a control module. According to the scheme, through organic integration of the crawler belt and self-adaptive wheel composite walking system, the modular high-speed cleaning wheel disc, the closed-loop negative-pressure scrap recycling device and the intelligent visual control system, the problems that traditional manual cleaning is high in risk and low in efficiency are solved; and the technical bottlenecks of an existing pipeline robot in the aspects of trafficability, stability, thoroughness in cleaning and environmental protection are overcome, high adaptability, high safety, high efficiency and zero secondary pollution of nuclear power station pipeline cleaning operation are comprehensively achieved, and reliable technical equipment support is provided for intelligent and unmanned operation and maintenance of a nuclear power station.

1. Name of the product in this design: Graphical User Interface for the Operation and Management of Medical Consumables for Electronic Devices. 2. Intended use of this design: for electronic devices; Intended use of the graphical user interface: for the operation and management of medical consumables for electronic devices. 3. The key design feature of this product is its graphical user interface. 4. The image or photograph that best illustrates the design's key features: the front view. 5. Human-computer interaction method of graphical user interface: The graphical user interface of this product is the user interface for running the vision control system software. The main change in the main view and various state change diagrams is that the corresponding interface state change diagram is presented according to the user's current control position. The main view is the main interface view of the software; enter your account and password, click the [Login] button to enter the changed state diagram 1; click the [Medical Consumables Operation Management] button in changed state diagram 1 to enter the changed state diagram 2; click the [Date Range] button in changed state diagram 2 to enter the changed state diagram 3.

This invention discloses a visual detection method for industrial production safety targets that integrates progressive feature pyramids. It acquires images of key areas in industrial settings, optimizes the target detection model through a lightweight feature extraction network and a dynamic threshold decision mechanism, introduces an adaptive feature pyramid network (AFPN) and an improved detection head structure, and enhances multi-scale feature fusion capabilities for detection. A hierarchical early warning system is constructed to achieve rapid response, accurate push notifications, and closed-loop management. A visual control system is configured to output the early warning results. This invention does not require additional dedicated detection equipment and is compatible with existing surveillance cameras or conventional industrial cameras in industrial settings. It can be implemented simply through algorithm optimization and architecture upgrades, effectively improving the feature capture efficiency for safety hazard targets of different distances and sizes, and significantly reducing category confusion and judgment errors in complex industrial scenarios.

The invention relates to the technical field of industrial robot visual control, and discloses a welding seam visual detection and robot coordinate conversion method and system based on multi-modal feature fusion, which can realize high-precision automatic detection and positioning of a welding seam in a complex industrial environment and accurately convert a detection result into a space coordinate which can be executed by a robot. According to the scheme, during modeling, images and teaching coordinates are collected, template blocks are cut, multi-modal features are pre-calculated, relevant data are stored after the physical distance is calculated, and indexes are updated; when the welding seam is detected, teaching coordinates are called to cut the reasoning block for detection, offset cutting and a model size backspacing mechanism are supplemented, then the welding seam is determined through multi-modal fusion similarity screening, and pixel coordinates are calculated; and finally, obtaining a depth value, carrying out distortion removal on the pixel coordinates to recover the three-dimensional coordinates of the camera, and carrying out homogeneous transformation to obtain the coordinates of the robot base and outputting the coordinates. The method is suitable for complex industrial environments with multiple welding seams, multiple interferences, high light reflection and the like, and the automation level and the welding precision of the welding robot can be remarkably improved.

The application discloses a kind of real-time video edge analysis methods of distributed visual controller, including, obtaining real-time video stream, and according to current computing resource and network state dynamic determination split strategy of deep neural network;Utilize the feature extraction part executed locally to video frame operation, and generate intermediate layer feature data;By combining pre-constructed channel sensitivity model and the feature activation intensity of current frame, the semantic importance parameter of each feature channel is determined;Based on the importance parameter, the semantic compression processing including channel screening and dynamic quantization is executed to feature data, and compressed feature data packet is generated;Finally, the data packet is sent to collaborative computing node to complete the final video analysis.The application can realize the adaptive allocation of end-edge computing power and intelligent data compression for machine vision tasks, while significantly reducing network bandwidth occupation, effectively guaranteeing the real-time performance and high precision of analysis task, improving the operation efficiency and robustness of the entire distributed system.

The invention discloses a rendering method for generating dynamic visual monitoring. The method comprises the following steps: acquiring initial data, executing baseline modeling, and generating a rendering cache, a layer dependency graph and an initial layout solution; in response to real-time data change, carrying out differential arrangement by utilizing a rendering cache and a layer dependency graph, defining a minimum influence domain and generating an execution sequence through dirty region propagation and neighborhood cutting of dependency perception; according to the minimum influence domain and the execution sequence, calculating stability measurement including indexes such as displacement increment and edge crossing increment, and performing decision submission based on a preset gating condition to obtain gated layout candidates; and generating a final presentation frame according to the gated layout candidates, and binding generation of the presentation frame and issuing of the equipment control instruction into a single transaction adopting a two-stage submission mechanism. According to the method, the rendering performance, the visual stability and the operation reliability of the dynamic monitoring system are improved by accurately defining the influence domain, introducing quantitative stability gating and guaranteeing the consistency of visual control transactions.

The invention relates to the technical field of industrial process control and digital visualization, in particular to a digital visualization control method for a food production line control system. The method comprises the following steps that the surface hardness parameter of food to be grabbed is obtained, and a contact sensitivity triggering threshold value is set according to the surface hardness parameter; the flexible air bag gripper is driven to descend to make contact with food, and signals collected by the flexible touch skin are converted into visual contact dynamic pressure distribution data; virtual package fitting degree simulation verification is carried out on the contact dynamic pressure distribution data and a contact sensitivity triggering threshold value in a digital visualization background, and a self-adaptive differential pressure adjusting signal is generated; inflation and deflation linkage of the independent miniature air chamber is triggered based on the self-adaptive differential pressure adjusting signal. According to the invention, the independent air chamber is driven by flexible tactile feedback to carry out self-adaptive differential pressure regulation and full-flow digital visual monitoring, so that lossless accurate grabbing of vulnerable food with irregular shapes and transparent quality management in the production process are realized.

This utility model relates to visual aids and educational equipment and is a desktop device for visually monitoring income and expenses. The device comprises a housing with functional compartments designed with cells for storing accounting elements, with the compartments equipped with markings and counters for recording the number of accounting elements. The housing is equipped with a mechanical balancing unit, including a longitudinal scale and a movable pointer, capable of movement along the scale and structurally connected to the income and expense sections, providing a display of their quantitative ratio. The technical result consists of providing a physical record of quantitative values ​​and a visual indication of the ratio of total income and expense values ​​without the use of computing technology.

The invention discloses a secondary cooling water distribution curve visual control method for a multi-strand billet continuous casting machine, which belongs to the technical field of metallurgy, and comprises the following steps: step S1, VBA programming and water distribution curve generation; s2, PLC program calculation is carried out; s3, PID adjustment is carried out; the method has the advantages that the set water distribution curve can be visually reflected, the matching accuracy is high, dynamic and accurate control can be conducted in combination with real-time parameters, and the diversified production requirements of the multi-strand billet continuous casting machine can be met. The problems that according to an existing control method, a set water distribution curve cannot be visually reflected, the corresponding cooling water amount cannot be accurately matched, dynamic accurate control combined with real-time parameters is lacked, a mechanism for evaluating and optimizing a water distribution program regularly based on production data is not formed, and then high-quality diversified production of the multi-flow billet continuous casting machine cannot be met are solved.

The application discloses an immersive customer engineering environment data display method and system, comprising the following steps: S1, an environment collection terminal collects engineering image data of a power supply scheme development stage site and returns the data to a workstation, and a BIM software is used to build an engineering three-dimensional model and engineering design drawings; S2, a management personnel collects expansion scheme environment image data according to customer engineering construction requirements, returns the data to the workstation, combines a classical scheme design library, drags a typical design on the engineering design drawings of the power supply scheme development stage, calls corresponding component attribute data, completes three-dimensional and two-dimensional standardized design of the engineering drawings, and performs graphic rendering on the three-dimensional model; and S3, an environment collection terminal collects environment image data, extracts features of the expansion scheme environment image, returns the data to the workstation, and the BIM software corrects the standardized engineering drawings into customer engineering completion design drawings. The scheme improves the visual control efficiency of engineering projects.

The present application relates to the technical field of deep learning machine vision control convex letter steel rolling, and relates to a convex letter steel rolling plate shape control method and system based on deep learning, which comprises the following steps: heating and descaling a blank, setting a first visual detection unit at the outlet of a heating furnace, triggering hydraulic correction when the centering deviation is greater than a first threshold value; setting a second visual detection unit at a rough rolling mill, extracting the residual area of the rolling piece surface oxide scale through an image segmentation algorithm, and detecting whether to trigger secondary descaling; setting a third visual detection unit at a medium rolling mill, adjusting the flow distribution of the cooling water of a finishing mill if the temperature distribution is uneven, and correcting the side pressure of the finishing mill vertical roll if the corner radius error exceeds a fourth threshold value; setting a fourth visual detection unit at the finishing mill, calculating the roll gap compensation amount and correcting the finishing mill reduction amount in real time when the filling rate is less than a fifth threshold value. The present application combines convex letter steel with deep learning and machine vision, and can make the quality of the manufactured convex letter steel better and the shape more accurate.