Patents
Literature
143 results about "Visual processing" patented technology
Efficacy Topic
Property
Owner
Technical Advancement
Application Domain
Technology Topic
Technology Field Word
Patent Country/Region
Patent Type
Patent Status
Application Year
Inventor
Visual processing is a term that is used to refer to the brain's ability to use and interpret visual information from the world around us. The process of converting light energy into a meaningful image is a complex process that is facilitated by numerous brain structures and higher level cognitive processes. On an anatomical level, light energy first enters the eye through the cornea, where the light is bent. After passing through the cornea, light passes through the pupil and then lens of the eye, where it is bent to a greater degree and focused upon the retina. The retina is where a group of light-sensing cells, called photoreceptors are located. There are two types of photoreceptors: rods and cones. Rods are sensitive to dim light and cones are better able to transduce bright light. Photoreceptors connect to bipolar cells, which induce action potentials in retinal ganglion cells. These retinal ganglion cells form a bundle at the optic disc, which is a part of the optic nerve. The two optic nerves from each eye meet at the optic chiasm, where nerve fibers from each nasal retina cross which results in the right half of each eye's visual field being represented in the left hemisphere and the left half of each eye's visual fields being represented in the right hemisphere. The optic tract then diverges into two visual pathways, the geniculostriate pathway and the tectopulvinar pathway, which send visual information to the visual cortex of the occipital lobe for higher level processing (Whishaw and Kolb, 2015).
Visual classification processing method and device based on large model and multi-modal data fusion
PendingCN121600326AGeometric image transformationCharacter and pattern recognitionProcessingAdaptive routing
The invention relates to the field of visual processing, and provides a visual classification processing method and device based on large model and multi-modal data fusion. The method comprises the following steps: inputting a to-be-classified input image and a corresponding category text description into a text encoder for multi-level feature extraction to obtain global text features and local text features; performing fine-grained cross-modal alignment on the local visual features and the local text features, calculating association weights between the image regions and the text phrases through a bidirectional cross attention mechanism, and generating aligned intermediate features; splicing and fusing the aligned middle features and the global visual features, and inhibiting background noise in a fusion result and reinforcing discriminative features in the fusion result through a feature mask algorithm in combination with the global text features to obtain multi-modal fusion features; and synchronously inputting the multi-modal fusion features into a multi-space classifier to generate respective classification results, and adaptively outputting an image classification result according to a confidence threshold in combination with a dynamic routing mechanism.
Owner:SUZHOU YINPO TECHNOLOGY DEVELOPMENT CO LTD
Road infrastructure abnormity identification method based on vision and AI intelligent agent
The invention discloses a road infrastructure anomaly recognition method based on vision and an AI agent. The method comprises the steps that a multi-agent cooperation framework with a task coordination agent as a core is constructed; a data input agent receives and preprocesses multiple paths of video streams; the visual processing agent fuses a visual model and a language model through a cross attention mechanism to realize deep fusion of image-text features; the knowledge base intelligent agent constructs a dynamic JSON Schema structured template based on a Neo4j graph database; and the data output agent calls the large language model to carry out data enhancement and structured processing, and the data is stored in a database after validity is verified by the knowledge graph. According to the method, through multi-agent cooperation, cross-modal feature fusion and a dynamic template technology, the problems of feature splitting, template stiffness and low calculation efficiency in a traditional scheme are solved, and the efficiency of road anomaly recognition is improved.
Owner:SOUTH CHINA UNIV OF TECH
Online monitoring system and method for ultrafast laser processing of micropores
PendingCN120734514AAnalysing solids using sonic/ultrasonic/infrasonic wavesMaterial analysis by optical meansVisual monitoringLaser processing
The invention discloses an online monitoring system and method for ultrafast laser processing of micropores. The online monitoring system comprises a visual monitoring unit, a sound wave monitoring unit and a judgment control module which are arranged in an ultrafast laser processing unit. The visual monitoring unit is used for obtaining a visual monitoring result and comprises a visual acquisition unit and a visual processing unit; the sound wave monitoring unit is used for obtaining a sound wave monitoring result and comprises a sound wave acquisition unit and a sound wave processing unit; and the judgment control module is used for comprehensively judging the visual monitoring result and the sound wave monitoring result so as to control the ultrafast laser processing unit. According to the technical scheme, online sound-vision composite monitoring can be carried out on the process of ultrafast laser micro-hole machining, and therefore the micro-hole machining efficiency and the penetration rate are improved.
Owner:HUAZHONG UNIV OF SCI & TECH
Target control method and device based on multi-modal information, equipment and medium
The invention relates to the technical field of artificial intelligence, can be applied to business scenes such as mechanical arm grabbing, financial science and technology and medical health, and discloses a target control method, device and equipment based on multi-modal information and a medium, and the method comprises the steps that a language instruction expressing an operation intention is obtained, and visual information of the environment where a target object is located is collected; semantic representation is generated through the language processing module, and visual features are generated through the visual processing module; inputting the visual features and the semantic representations into an action generation module to generate an action sequence; and a joint angle sequence and a joint angle value are determined based on the action sequence, and the mechanical arm is controlled to execute grabbing operation. According to the method, the accurate grabbing action sequence is generated by fusing language and visual information, the movement of the mechanical arm is controlled by combining the joint angle sequence and the accurate joint angle value, and smooth track movement and accurate grabbing are achieved. The adaptive capacity and the industrial-grade grabbing precision in the complex environment are improved, and the production efficiency and the grabbing success rate are effectively improved.
Owner:PING AN TECH (SHENZHEN) CO LTD
Automatic saddle welding device based on visual processing and using method thereof
PendingCN121670221AWelding/cutting auxillary devicesAuxillary welding devicesVision basedControl engineering
The invention discloses an automatic saddle welding device based on visual processing and a using method thereof, and relates to the field of saddle welding.The device comprises a conveyor, a clamping mechanism, a lifting mechanism, an electric rotating disc, a welding mechanical arm and a visual positioning device.A supporting table at the top of a base can move along a guide rail D and is matched with production line procedure arrangement; the clamping mechanism drives clamping plates to move oppositely through a screw A, flexible clamping is achieved in combination with a spring, workpiece damage is avoided, and the posture stability is guaranteed. The core logic of clamping first and then positioning is adopted, the visual positioning device accurately extracts welding seam feature point coordinates through a two-dimensional image acquisition module, a three-dimensional contour acquisition module and a light supplementing adjusting module, and the welding seam feature point coordinates are accurately obtained; positioning data are provided for a welding mechanical arm, the problem of welding deviation caused by workpiece transfer displacement, clamping posture adjustment and coordinate system difference in the prior art is effectively solved, and the defects of weld joint deviation, incomplete penetration and the like are overcome.
Owner:NANJING XUNSI INTELLIGENT MANUFACTURING TECHNOLOGY CO LTD
Power transmission pole foundation pit scanning sonar detection device, system and method based on unmanned aerial vehicle
PendingCN121114210AAnalysing solids using sonic/ultrasonic/infrasonic wavesDrawing from basic elementsSonarTerminal equipment
The invention discloses a power transmission pole foundation pit scanning sonar detection device, system and method based on an unmanned aerial vehicle, and the method comprises the steps: receiving a detection task instruction inputted by terminal equipment, carrying out the initialization and path planning according to the received detection task instruction, and carrying out the calibration; setting an automatic program; generating a defect identification result and an execution signal; performing visualization processing on the acquisition process and the execution process to generate a detection report and a defect distribution diagram; and carrying out quality analysis and classification processing. In the implementation process of the technical scheme, the detection task instruction is input through the terminal equipment, accurate positioning and automatic detection of the power transmission pole foundation pit are achieved, the execution signal is generated according to the defect recognition result, the unmanned aerial vehicle automatically executes the corresponding defect processing operation, and therefore the efficiency and accuracy of foundation pit detection are improved; and meanwhile, the collected data is visually displayed through the visual processing module, and a visual detection report and a defect distribution diagram are generated.
Owner:JINHUA BADA GRP CO LTD
Paper tray production quality detection method and system based on visual processing
ActiveCN120496055AOptically investigating flaws/contaminationManufacturing computing systemsAlgorithmVision based
The invention discloses a paper tray production quality detection method and system based on visual processing, and belongs to the technical field of quality detection, and the method comprises the steps: collecting image and motion data, carrying out the alignment according to a time sequence, carrying out the risk classification, marking a high-risk region, and constructing a stress prediction distribution diagram. Calculating an angle residual error according to an existing image and a stress prediction distribution diagram, calculating the angle residual error diagram to obtain a regional average residual error value, taking the regional average residual error value as a weighting coefficient and a deviation distribution index to generate speed and angular speed regulation factors, and finally generating a mold execution instruction of the next period based on an adjustment strategy. According to the detection method, process modeling and structure perception are combined, the defects of a traditional paper support defect detection means based on a static image in the aspects of time hysteresis, structure invisibility, control inaccessibility and the like are effectively overcome, and the intelligent level and the detection coverage rate of paper support quality detection are remarkably improved.
Owner:QINGYUAN KEDING ELECTRICAL EQUIP CO LTD +2
Novel reading disorder assessment and intelligent correction method based on EOG
The invention relates to a novel reading disorder assessment and intelligent correction method based on EOG. The novel reading disorder assessment and intelligent correction method comprises the following steps: S1, respectively collecting eye movement signal data of a reading disorder child and a normal reading child in a reading process by adopting eye movement collection equipment; s2, filtering the original signal to generate an eye movement diagram, and analyzing transverse and longitudinal movement tracks and directions of eyes; s3, comparing the eye movement images of the reading disorder children and the normal children, and extracting key eye movement characteristic parameters of the reading disorder children and the normal children; s4, analyzing eye movement abnormal characteristics of frequent review of the individual; and S5, assessing the reading disorder through the reading behavior, and assessing the understanding degree of the text segment by using the text-related picture. The intelligent correction module is started to improve the reading effect. According to the method, the motion state of eyeballs, individual attention, visual processing speed, cognitive load and the like are reflected through the change of the electro-oculogram potential, a non-intrusive, real-time and quantitative evaluation and correction scheme is provided, and the evaluation and correction effect of the reading disorder can be remarkably improved.
Owner:ZHEJIANG MEDICAL COLLEGE
Super-resolution image real-time processing method and device fusing attention mechanism
The invention relates to the field of visual processing, and provides a super-resolution image real-time processing method and device fusing an attention mechanism. The method comprises the following steps: acquiring a to-be-processed low-resolution image; performing initial feature coding on the low-resolution image to obtain a shallow feature map; performing adaptive modeling on the feature contribution degree of each spatial position and each feature channel in the shallow feature map in combination with a mixed attention mechanism, and generating an attention-enhanced depth feature map; converting a feature map channel number of the depth feature map into a spatial resolution, converting image feature information in the depth feature map according to a feature contribution degree and re-projecting the image feature information to a target image space corresponding to the spatial resolution, and generating a high-resolution feature map corresponding to a target magnification in the target image space through a pixel rearrangement mode; and reconstructing the high-resolution feature map into a target high-resolution image. Efficient real-time processing of the super-resolution image is achieved, and the image reconstruction definition and the detail reduction degree are improved.
Owner:SUZHOU YINPO TECHNOLOGY DEVELOPMENT CO LTD
Underwater robot deep learning adaptive data processing method and system
PendingCN120783202ACharacter and pattern recognitionNeural learning methodsData acquisition3d localization
The invention discloses an underwater robot deep learning adaptive data processing method and system, and relates to the technical field of deep learning. Sonar data, a visual image and a robot motion state of an underwater environment are collected; enhancing the definition of a visual image by using a convolutional neural network, marking autonomous moving elements, and constructing a two-dimensional coordinate system to complete preliminary positioning; the element depth is calculated based on the pixel value and the time delay of the sonar data, and the two-dimensional coordinates are fused to construct a three-dimensional coordinate system; and unifying the track in the three-dimensional space, calculating the distance between the elements and the robot, and planning the moving route and speed. The system comprises a data acquisition module, a visual processing module, an acoustic fusion module and a path planning module, and realizes multi-source data integration, three-dimensional positioning and path planning. Through visual and acoustic data fusion and a deep learning model, the underwater environment stereoscopic perception precision and the path planning efficiency are improved, and the adaptability and the operation safety of the underwater robot in a complex water area are enhanced.
Owner:CHANGZHOU ZHIGAN TECH CO LTD
Traffic condition prediction method and device and computer program product
The embodiment of the invention discloses a traffic condition prediction method and device and a computer program product. According to the main technical scheme, the method comprises the steps of obtaining trajectory data of vehicles in a target area; performing visualization processing on the trajectory data to obtain visual data; the traffic condition of the target area can be obtained by further combining effective traffic information extracted from at least one of traffic intelligence data, user report data and environmental data based on visual data by utilizing a visual language model. The traffic condition comprises at least one of the congestion degree, the occurrence position of the traffic event, the occurrence reason and the congestion trend information. According to the mode, the comprehensive and accurate traffic condition of the target area can be obtained by utilizing the visual language model based on the visual data.
Owner:BEIJING AUTONAVI YUNMAP TECH CO LTD
Power consumption management method and system for electric energy metering box
The invention relates to the technical field of electric energy metering, and discloses a power consumption management method and system of an electric energy metering box. The method comprises the following steps: acquiring an original electric signal containing a plurality of electric parameters, and generating a steady-state and transient-state data block through filtering noise reduction, variable-point detection and time domain segmentation processing; based on the data block, obtaining a standardized feature vector through feature extraction, feature sorting and normalization processing; carrying out load identification and time alignment on the feature vector by utilizing transfer learning and confidence judgment, and generating a load type identifier with a timestamp; the load access time is further analyzed and standardized, and structured intermediate data is generated through space-time alignment and data fusion; and finally, a power consumption management report is formed through visualization processing. According to the invention, the precision and reliability of non-intrusive load monitoring can be effectively improved, fine-grained and automatic analysis and visual feedback of power consumption behaviors of users are realized, and the method has the beneficial effects of high data processing efficiency, high identification accuracy and visual and convenient management.
Owner:BAODING XINDI ELECTRIC TECH CO LTD
Foundation engineering sediment detector based on physical sensor
PendingCN120211333AFoundation testingIn situ soil foundationStructural engineeringStandard penetration test
The invention relates to the field of foundation engineering detection, in particular to a foundation engineering sediment detector based on a physical sensor, which comprises an automatic coupling agent filling control module, a pile foundation sediment thickness analysis module, a foundation settlement early warning module, a standard penetration test module, a data association module and a comprehensive evaluation module, according to the system, coupling agent filling abnormal data and pile foundation hole sediment thickness abnormal data are analyzed and screened, a change curve of foundation settlement along with time is drawn, a threshold value is set to trigger early warning, and the penetration number and depth data are collected to determine the compactness grade of a foundation soil layer, so that a data association relation table is constructed, and visualization processing and presentation are carried out; meanwhile, the comprehensive evaluation score of the foundation is comprehensively calculated, the stability degree grade of the foundation is determined and fed back, the filling speed of the coupling agent is conveniently adjusted, the accuracy of pile foundation sediment thickness detection data is ensured, and the efficiency and practicability of foundation settlement monitoring and the standard penetrometer are improved.
Owner:SUZHOU TRAFFIC ENG TESTING CENT CO LTD
Tunnel deformation detection method based on convolutional neural network and machine vision sensor
The invention relates to the field of tunnel engineering safety monitoring, in particular to a tunnel deformation detection method based on a convolutional neural network and a machine vision sensor, and the method comprises the steps: collecting annular deformation data, longitudinal deformation data and cross section displacement inclination angle data through the machine vision sensor arranged in a tunnel; preprocessing the annular deformation data and the longitudinal deformation data by using a convolutional neural network; the cross section displacement inclination angle data are converted into discretized matrix data, and visualization processing is carried out based on the differential geometry theory; constructing a tunnel deformation monitoring model and carrying out simulation measurement and calculation; according to the method, the differential geometry theory and the convolutional neural network are deeply fused, the cross section of the tunnel is regarded as a two-dimensional Riemannian manifold embedded into a three-dimensional Euclidean space, the deformation characteristics of a complex curved surface are accurately described, and the deformation speed of the tunnel is determined. All-directional and high-precision monitoring and prediction of tunnel deformation are realized.
Owner:BEIJING DITIE ARCHITECTURE INSTALL ENG CO
Chassis prediction control system and method based on binocular vision and multi-domain cooperation
The invention relates to the technical field of vehicle electric control suspensions, in particular to a chassis prediction control system and method based on binocular vision and multi-domain cooperation, and constructs the chassis prediction control system based on binocular vision and multi-domain cooperation based on an ADAS binocular camera. IMU data of a binocular camera is obtained by using a data acquisition module of the chassis prediction control system, visual observation data of a vehicle body is obtained by using a binocular visual processing module, and then the IMU data and the visual observation data are fused and converted into a three-dimensional image by using modules such as an extended Kalman filtering fusion module and a motion conversion module. The sprung mass vertical acceleration and sprung mass vertical speed of the vehicle four-corner suspension provide accurate suspension four-corner vertical motion state input for an electric control suspension controller, and the accuracy and real-time performance of the electric control suspension for vehicle body posture control are guaranteed. Meanwhile, the problems that a special IMU is difficult to arrange, high in cost and insufficient in precision and IMU resources of an ADAS system are wasted are solved.
Owner:辰致科技有限公司
Damage three-dimensional reconstruction and full-time-history quantitative analysis method based on two-dimensional axisymmetric finite element model
The invention discloses a damage three-dimensional reconstruction and full-time-history quantitative analysis method based on a two-dimensional axisymmetric finite element model, and the method comprises the steps: constructing the axisymmetric finite element model, importing the axisymmetric finite element model into LS-Dyna software, carrying out the numerical calculation, analyzing a K file, and extracting key information. And screening a specified unit, preprocessing data, calculating the centroid position and volume, and realizing accurate mapping from two dimensions to three dimensions. Mechanical response parameter time history data are extracted, binarization processing and peak value extraction are carried out, damage parameters are calculated, damage time history data are subjected to visualization processing, and a high-resolution image is output. The method solves the problems of lack of spatial-temporal evolution analysis capability, blank extraction of damaged geometric parameters, insufficient cloud picture resolution and the like in the prior art, has wide applicability and practical application value, and can provide powerful support for explosion impact dynamics and penetration damage evaluation.
Owner:NANJING UNIV OF SCI & TECH
Radial ultrasonic roller device for cleaning wire rod structure
ActiveCN120038163ADrying solid materials without heatMechanical vibrations separationRotational axisWire rod
The invention discloses a radial ultrasonic roller device for cleaning a wire structure, which comprises a roller, an ultrasonic generator, a visual processing system, a high-speed camera, a tensioning wheel, a mounted bearing, a rotating shaft, a connecting wheel, a via hole conductive slip ring, a rotation stopping fixer, a bracket and a stand column, the roller is fixedly installed on a rotating shaft, and the two ends of the rotating shaft are rotationally installed on a support. Through the synergistic effect of ultrasonic vibration and a roller structure, rapid atomization and separation of liquid drops are achieved, the wire surface drying efficiency is remarkably improved, and meanwhile the problems of high energy consumption or low efficiency in a traditional method are solved; ultrasonic acts on the wire rod through the roller and does not make direct contact with the wire rod, the wire rod is prevented from being damaged due to the direct action of the ultrasonic, the wire passing groove is formed in the roller, the wire rod moves through the wire passing groove to be clamped, and deviation of the wire rod in the moving process is avoided.
Owner:BEIJING UNIV OF POSTS & TELECOMM
Double-station solder ball welding machine correction system based on machine vision and multi-axis cooperation
The invention discloses a double-station solder ball welding machine correction system based on machine vision and multi-axis cooperation, and belongs to the field of welding equipment correction. A double-station solder ball welding machine correction system based on machine vision and multi-axis cooperation comprises an image acquisition unit, a visual processing unit, a multi-axis control unit and a cooperative scheduling unit. The problems that in existing double-station welding equipment, due to the fact that a single-station vision scheme is simply copied, the equipment takt is slow, the productivity is limited, and comprehensive compensation for dynamic errors and optical environment differences under multi-station collaborative operation is lacked are solved. Compared with the prior art, the double-station equipment has the advantages that overlapped execution of visual correction and welding operation is achieved, and feed-forward and online compensation mechanisms for dynamic errors such as thermal deformation and vibration are integrated, so that high-precision positioning is guaranteed, and the production efficiency and the operation reliability of the double-station equipment are remarkably improved.
Owner:SHENZHEN INFEASANT TECH CO LTD
A mindfulness prediction-intervention-feedback system and control method
The application discloses a mindfulness prediction-intervention-feedback system and a control method, relates to the technical field of mindfulness training systems, and comprises a wearable user interaction device and sequentially connected user management module, mindfulness intervention training module, dynamic prediction module, intervention effect evaluation module and feedback interaction module; the initial mindfulness level, depression level and anxiety level of a user are determined; based on the depression level and anxiety level of the user, a target training scheme of the user is determined, and the user is prompted to perform mindfulness training according to the target training scheme; the target training scheme is adjusted based on an emotional state factor; the real-time mindfulness level of the user and the completion progress of the target training scheme by the user are determined according to a preset time interval, a mindfulness training report is generated in combination with the initial mindfulness level, and visual processing is performed; the mindfulness training method is selected by evaluating the mindfulness level of the user, and the rationality of the mindfulness training scheme can be improved.
Owner:GUANGDONG UNIVERSITY OF FOREIGN STUDIES
Slope stability evaluation method and system under heavy rainfall condition
The invention provides a slope stability evaluation method and system under a heavy rainfall condition, and the method comprises the steps: collecting a slope image data sample, extracting a two-dimensional slope section and key geometric feature parameters of the slope image data sample through employing a three-dimensional space modeling technology, and constructing a data set through employing a two-dimensional finite difference numerical simulation technology; constructing a hybrid model of an intelligent optimization algorithm based on supervised learning and machine learning regression, and training the hybrid model by using the optimized regression model hyper-parameters and the data set to obtain a slope stability evaluation model; inputting a to-be-detected slope image of a target area into the slope stability evaluation model to obtain a slope safety stability coefficient result; and carrying out interpretability analysis and visualization processing on the slope safety stability coefficient result to obtain a slope safety stability coefficient correlation parameter analysis result and a visualization display result.
Owner:QINGHAI HUANGHE HYDROPOWER DEVELOPMENT CO LTD +2
Intelligent investment advisor interaction and asset allocation visual decision-making system based on augmented reality
The invention relates to the technical field of intelligent investment advisers, and discloses an intelligent investment adviser interaction and asset allocation visual decision-making system based on augmented reality, and the system comprises an augmented reality display device; the user interaction module is used for capturing an input instruction of a user; a data acquisition module; intelligent investment advisor and commissioning; a visualization processing unit; a decision execution module; a system control center; a user identity authentication module; a data storage module; a network communication module; and a calculation acceleration module. According to the intelligent investment adviser interaction and asset allocation visual decision-making system based on augmented reality, by integrating augmented reality display equipment and a multi-dimensional visual processing unit, an abstract asset allocation scheme, a market trend and risk information are overlaid in a real environment in visual forms such as a three-dimensional hologram and a spectrogram; the understanding efficiency and decision immersion of the user on the complex financial data are remarkably improved, and the problems that traditional two-dimensional interface information is not visual in presentation and poor in interaction experience are solved.
Owner:UNIV OF SCI & TECH OF CHINA
Patient somatosensory monitoring system, method, device and equipment
The invention relates to a patient somatosensory monitoring system, method, device and equipment. The system comprises a magnetic resonance imaging device used for performing magnetic resonance scanning on a patient; the sensor is used for collecting local somatosensory data of the patient during magnetic resonance scanning in the magnetic resonance imaging equipment; the visual processing unit is used for receiving the local somatosensory data collected by the sensor, and predicting to obtain global somatosensory data according to the local somatosensory data; wherein the type of the local somatosensory data is the same as or different from that of the global somatosensory data; and the display unit is used for receiving the global somatosensory data sent by the visualization processing unit and displaying the global somatosensory data. By adopting the scheme, the arrangement difficulty of the sensor can be reduced.
Owner:SHANGHAI UNITED IMAGING RES INST OF INTELLIGENT IMAGING
Method for judging compaction degree of working face old roadway based on SF6 air leakage measurement
PendingCN121384931ADetection of fluid at leakage pointMining devicesMining engineeringStructural engineering
The invention discloses a working face old roadway compaction degree evaluation method based on SF6 air leakage measurement, and relates to the field of compaction evaluation, and the method comprises the following steps: obtaining coal mine basic information, and setting a coal mine working face three-dimensional model; setting a corresponding release point and a plurality of detection points according to the coal mine working face three-dimensional model; setting the required release volume according to the release point position information, releasing the corresponding tracer gas, and obtaining the detection information of the tracer gas at each detection point position; carrying out visualization processing on the obtained detection information of the tracer gas; the visual processing results of the detection information corresponding to all the detection points are compared and analyzed, and corresponding air leakage measurement information is obtained; ventilation route information in the mine is obtained according to the air leakage measurement information obtained by all the detection points, analysis processing is conducted according to the ventilation route information, and compaction degree information corresponding to the old roadway information is obtained; the safety of the working face old roadway is improved.
Owner:HUAI-ZHEJIANG COAL & ELECTRICITY CO LTD GUBEI COAL MINE +2
A real-time monitoring and early warning method and system for key impurities based on optoelectronic impurity removal of cut tobacco
The present invention relates to the field of tobacco impurity removal, and proposes a real-time monitoring and early warning method and system for key impurities based on optoelectronic impurity removal of cut tobacco. First, scan the tobacco leaves and collect images to obtain high-quality tobacco leaf images. Second, determine whether there are impurities in the tobacco leaf images. If there are, they are used as impurity images. Third, cache continuous multiple segments of impurity images, splice the continuous multiple segments of impurity images into a complete impurity image through timestamps, and upload it to the upper industrial control computer. Fourth, the upper industrial control computer monitors the complete impurity image in real time. The image is displayed on the statistical dashboard, and key impurities are identified based on the AI classification algorithm, and it is judged whether the impurity belongs to a key impurity. If it belongs to a key impurity, an audible and visual alarm is issued. The present invention realizes the real-time monitoring and early warning of key impurities, and solves the technical problem that the existing technology is in a black box state for tobacco impurity removal and cannot perform visual processing on impurity information.
Owner:QILIN REDRYING FACTORY YUNNAN TOBACCO REDRYING +2
Method and apparatus for generating key points
The disclosure discloses a key point generation method and device, relates to the technical field of computers, and particularly relates to the technical field of visual processing. A specific implementation scheme is as follows: first, a target human body model corresponding to a target object is acquired; then, position information of a vertex in the target human body model is determined; finally, the position information of the vertex in the target human body model is input into a key point generation model, a human body key point corresponding to the target object is generated, the target human body model is identified through the constructed key point generation model, the human body key point corresponding to the target human body model is determined, and thus the human body key point corresponding to the target object is determined. The human body key point can be generated based on a model surface vertex, is suitable for various flexible human bodies with various postures and muscle non-rigid changes, and the accuracy and flexibility of the human body key point are improved.
Owner:BEIJING BAIDU NETCOM SCI & TECH CO LTD
Lamination thickness detection system and method for transformer core production line
PendingCN120991728AUsing optical meansApparatus for force/torque/work measurementProduction lineTransformer
The invention discloses a thickness detection system and method, belongs to the technical field of transformer core lamination thickness detection, and particularly relates to a lamination thickness detection system and method for a transformer core production line, and the system comprises a beat monitoring module, a lamination thickness detection module, an upper computer, and a wireless communication module. According to the invention, the lamination thickness data of each step can be effectively detected in the automatic iron core lamination process. Through high-speed visual processing, a 3D laser sensor and a mobile robot technology, a detection position is rapidly and accurately positioned, the production takt is followed, and a production line is helped to find and process abnormal data in time, so that the production efficiency and the product quality are improved, and an important support is provided for automatic and intelligent development of a production process.
Owner:JIANGSU SENLAN INTELLIGENCE SYST CO LTD
Bearing fault diagnosis method based on one-dimensional convolutional neural network
The invention relates to a bearing fault diagnosis method based on a one-dimensional convolutional neural network, and relates to the technical field of bearing fault diagnosis. The method comprises the steps that a bearing working data set is acquired, wherein the bearing working data set is a data set acquired in the working process of a target bearing; the bearing working data set is input into a bearing fault diagnosis model, a bearing fault diagnosis result corresponding to the target bearing is output, and the bearing fault diagnosis model is an artificial intelligence model based on a one-dimensional convolutional neural network; and carrying out visualization processing on the bearing fault diagnosis result. After a bearing working data set is obtained, a bearing diagnosis result corresponding to the bearing working data set is generated through an artificial intelligence model based on a one-dimensional convolutional neural network, and a bearing fault diagnosis result is obtained after visual result processing. And an artificial intelligence technology and a visual processing process are combined, so that the processing result of the bearing working data set is efficient and accurate.
Owner:WUXI WEIMING INTELLIGENT TECH CO LTD
Guardrail topological graph generation system based on guardrail connection identification
The invention relates to the field of guardrail topology generation, and particularly discloses a guardrail topological graph generation system based on guardrail connection identification, which comprises the following steps: firstly, acquiring an original signal flow generated by connection or disconnection operation in real time through a sensor deployed at a guardrail connection port; the signals are then parsed into a structured event list. By analyzing the time sequence of the events, the active connection event of one guardrail is matched with the passive connection event of the other guardrail in an extremely short time window, so that the connection relationship between the guardrails is accurately identified. Once the connection relation is confirmed, nodes representing guardrails and edges representing connection are generated on the basis, and a logic topological graph is formed. And finally, performing visualization processing on the logic diagram to generate a visual layout form image. Therefore, the real-time and remote visual supervision of the guardrail topology is realized, and the construction management efficiency and the safety response speed are effectively improved.
Owner:ZHEJIANG HAIYAN POWER SYST RESOURCES ENVIRONMENTAL TECH
Multi-modal image defogging method and device, electronic equipment and storage medium
The invention provides a multi-modal image defogging method and device, electronic equipment and a storage medium, and relates to the technical field of image processing, and the method comprises the steps: carrying out the feature extraction of an RGB image and radar point cloud data, and obtaining RGB image features and radar point cloud features; inputting into a first parallel convolution module and a second parallel convolution module for convolution processing to respectively obtain a first feature fusion result and a second feature fusion result; performing feature fusion through the first feature fusion result and the second feature fusion result to obtain a multi-modal image fusion result; based on a visual processing improved algorithm, a multi-modal image fusion result is input into an image defogging model to obtain a fused defogged image, the image defogging model comprises a target encoder, a target decoder and a feature enhancement module, and the feature enhancement module is constructed based on Vision Transform. According to the invention, the image defogging precision is improved.
Owner:CHINA UNIV OF GEOSCIENCES (WUHAN)
An interactive control method and system for an industrial vision system
PendingCN122174821ANatural language data processingKnowledge representationData acquisitionProcessing
The application discloses an interactive control method and system of an industrial vision system, and relates to the technical field of data analysis.The method comprises the following steps: preprocessing user request information; performing intention analysis on the preprocessed user request information based on a structured knowledge base; matching a corresponding core function module based on intention analysis information to determine instruction information, and performing standardization processing on the instruction information based on a parameter name mapping; issuing the standardized instruction information to the industrial vision system for execution based on a system interfacing layer; in the execution process, acquiring execution state data and execution effect data of the industrial vision system in real time for visual processing, and storing the visual data and the intention analysis information into a data storage layer.The application realizes full-process automatic execution of data collection-analysis-visual data generation, and realizes intelligent interaction of the industrial vision system.
Owner:SHENZHEN WEINER HUISHI TECHNOLOGY CO LTD