458 results about "Camera array" patented technology

The invention relates to the technical field of image processing, and discloses a stranded wire production quality monitoring method and system based on machine vision. The method comprises the following steps: acquiring the surface of a stranded wire at multiple angles through a high-speed camera array to obtain an original image; filtering high reflection and noise to obtain a feature map; performing edge detection, segmenting and identifying defects, and generating a position mapping table; quantitatively calculating defect parameters to form a quality score; performing correlation analysis to adjust process parameters; and constructing a defect-process database to realize quality tracing early warning. According to the invention, full-process automatic control from defect identification, quantitative evaluation to process parameter optimization is realized, the defect-process relational database is constructed, and the tracing and early warning functions of production quality are supported, so that the quality stability and the intelligent level of stranded wire production are improved.

The invention relates to the technical field of industrial automatic detection and intelligent operation and maintenance, in particular to a machine vision and big data fused linear slide rail defect identification and prediction system, which comprises a machine vision acquisition module, a time sequence feature extraction module, a big data analysis platform, a defect prediction and decision module and the like. The machine vision acquisition module acquires a slide rail image through an annular camera array and a multi-angle light source, and the acquisition quality is ensured through motion compensation; the time sequence feature extraction module utilizes a space-time attention model to mine defect features; the big data analysis platform is associated with the multi-source data to construct a defect evolution graph; the defect prediction and decision-making module outputs residual life prediction and maintenance instructions based on double-target reinforcement learning, and the dynamic model updating unit optimizes the model according to errors. According to the method, high-precision identification, prospective prediction and intelligent decision making of the defects of the linear sliding rail are realized, the equipment reliability can be effectively improved, the operation and maintenance cost is reduced, and the method is suitable for a complex industrial environment and has wide application value.

The invention discloses an intelligent emergency and management and control method for a stadium, and belongs to the technical field of stadium safety management and control. According to the method, video stream data are collected through a monitoring camera array, and coverage areas of all cameras in the array have 15%-20% of overlapping areas; inputting the data into a pre-trained people density identification model to generate a real-time people flow distribution thermodynamic diagram; the retention index of each partition is calculated, and an emergency evacuation mode is activated in combination with a preset safety threshold value; in the emergency mode, the infrared sensor group monitors the instantaneous pedestrian flow speed, and an optimal evacuation path is generated by a dynamic path planning algorithm; the path is converted into a control instruction to drive an electromagnetic access control gate, emergency lighting and an electronic guide identifier, feedback data triggers a path re-calculation mechanism, and a two-dimensional code scanning device counts the number of evacuated people so as to start manual re-checking. Real-time situation awareness, dynamic risk assessment and intelligent evacuation guidance of the stadium are achieved, emergency response and evacuation efficiency is effectively improved, and crowd safety is guaranteed.

The invention relates to the technical field of vision, in particular to an object monitoring method based on a multi-camera joint calibration technology and a monitoring camera thereof. The method comprises the following steps: driving all cameras to synchronously shoot a calibration reference object with known geometric characteristics, and solving internal and external parameters of each camera and an accurate space pose relationship between the internal and external parameters by utilizing shot images to form a joint space relationship chain; controlling the camera array to synchronously shoot a target object, and correspondingly converting the two-dimensional image points acquired by the cameras into spatial point coordinates in a unified three-dimensional coordinate system by using the joint spatial relation chain to generate three-dimensional point cloud data of the target object; and processing the generated three-dimensional point cloud data, extracting key geometric features of the target object in real time, performing comparative analysis on the extracted features and a preset standard or a historical state, and outputting a state monitoring result of the target object. According to the invention, the influence of temperature drift and mechanical vibration on the measurement precision is effectively suppressed, and the stability of long-term monitoring of an industrial field is guaranteed.

The invention relates to the field of industrial cameras, and particularly provides a panoramic splicing type industrial camera device for a complex scene and a splicing method, and the panoramic splicing type industrial camera device comprises a multi-camera array, a synchronous control module, an image processing module, a calibration module and a communication module. The multi-camera array is distributed in an annular or matrix mode. The synchronous control module realizes sub-microsecond synchronization (error is less than 0.8 microsecond) based on an LVDS hardware trigger architecture, and eliminates splicing dislocation of a high-speed moving target. The image processing module carries an FPGA hardening assembly line and completes image distortion correction, optimization ORB feature matching and multi-band fusion in real time. The calibration module integrates a multi-mode sensing unit, and dynamically adjusts exposure, focal length and white balance through improved Kalman filtering and a deep reinforcement learning decision algorithm. In a strong interference environment, the panoramic stitching precision reaches 0.01 mm, the defect detection sensitivity is 30 [mu] m, the power consumption is only 12 W, and the system is suitable for various industrial scenes.

The invention relates to the technical field of panoramic image construction, in particular to a panoramic image reconstruction method and system based on multi-angle imaging. The method comprises the following steps: collecting a multi-angle original image based on a distributed multi-camera array, carrying out adaptive filtering denoising and adaptive panoramic imaging adjustment, and constructing a multi-angle imaging geometric constraint network; performing multi-view semantic information deviation elimination based on a multi-angle imaging geometric constraint network, and performing global semantic feature fusion to obtain a unified semantic space representation framework; identifying illumination feature information of different visual angles, performing multi-angle illumination corresponding compensation on the multi-angle original image, performing image semantic distortion correction based on a unified semantic space representation framework, and constructing a multi-angle illumination compensation image; and performing multi-scale texture structure analysis on the multi-angle illumination compensation image to generate a high-fidelity texture fusion image. According to the invention, a natural and seamless panoramic image is provided, a panoramic scene is perfectly presented, and the immersive visual experience of a user is improved.

The invention relates to the technical field of image compensation, in particular to a reflection compensation method based on image processing, and provides the following scheme: acquiring an environment panoramic image by using a panoramic camera to establish a scene global coordinate system, and acquiring a multi-view original image in combination with a camera array arranged in the circumferential direction of an object; determining a rotation symmetry axis according to the multi-view contour features, reconstructing a three-dimensional geometric body, and calculating normal distribution and curvature change; generating a prediction image based on diffuse reflection and specular reflection hypothesis in the surface expansion domain through virtual visual angle disturbance, identifying a reflection region according to color and gradient consistency, and generating a reflection mask; and performing texture reconstruction on the reflective area by using geometric registration and multi-view compensation, and finally performing splicing and fusion to obtain a non-reflective high-fidelity panoramic image. The method does not need to change the field illumination condition, and can achieve the precise recognition and compensation of the complex curved surface reflection in the cultural relic in-situ collection environment.

The invention relates to the technical field of industrial manufacturing digital management, in particular to a machine vision production line performance evaluation and optimization management system, which comprises a data acquisition module for triggering a high-speed industrial camera array, a vibration sensor and an RFID reader through a central synchronous controller to synchronously acquire product images, equipment operation and material circulation data; the data processing and fusion module extracts product quality features based on CNN, and fuses multi-modal data through time sequence alignment normalization and an attention mechanism; the dynamic efficiency evaluation module calculates OEE, FPY and a production line balance rate in real time by means of a deep neural network; the optimization strategy generation module is used for reinforcing the learning agent to output optimization instructions such as equipment parameter adjustment; and the control execution module converts the instruction into an industrial protocol format, issues the instruction to the PLC, and verifies the effect to form a closed loop. According to the method, the data relevance and the evaluation real-time performance are improved, the dynamic state of the adaptive production line is optimized, and the efficiency improvement is facilitated.

The invention relates to the technical field of spine rehabilitation training, and discloses a computer vision recognition and guidance system for spine rehabilitation training actions. The system comprises a data acquisition module for acquiring user spine motion three-dimensional dynamic image data by using a multi-camera array; the action recognition module tracks bone joint points based on an improved algorithm to generate a real-time action sequence; the posture evaluation module is used for extracting spine curvature and joint angle features through a multi-scale convolutional neural network; the guidance instruction generation module inputs the features into a reinforcement learning model to generate a personalized instruction; and the feedback execution module constructs a fuzzy self-adaptive control model to correct the motion trail of the spine. According to the system, training actions can be accurately recognized, personalized guidance is provided, real-time correction is achieved, the spine rehabilitation training effect is effectively improved, subjectivity and errors of manual evaluation are reduced, and rehabilitation requirements of different patients are met.

The invention discloses a mechatronics product detection system based on an artificial intelligence image recognition technology, and relates to the technical field of artificial intelligence and mechatronics, and the mechatronics product detection system comprises an image acquisition module, a preprocessing module, a feature extraction module, a defect recognition module, a data management module, a result output module and a system control module, the preprocessing module has the functions of adaptive zooming and intelligent obstacle avoidance, and processes images by using a deep learning algorithm and a quantum filtering technology; a deep convolutional neural network is used for feature extraction, SVM and CNN are combined for defect recognition, a meta-learning mechanism is introduced, a distributed database and federated learning are adopted for data management, AR navigation and gesture interaction is used for result output, and a block chain intelligent contract is used for system control to deploy resources. The method has obvious advantages and high detection precision, and can accurately identify tiny and hidden defects; efficiency is greatly improved, and edge and cloud computing are combined to reduce processing time; operation is convenient, VR, AR and voice interaction are used, and the maintenance efficiency is improved.

The invention relates to the technical field of security management and control, in particular to an intelligent security dynamic identification management and control system, which comprises a multi-mode sensing module, a high-definition camera array, a millimeter-wave radar, an infrared sensor, an RFID reader-writer and a data processing module, and is used for receiving data of the multi-mode sensing module, carrying out time-space alignment and data fusion processing, and sending the data to the multi-mode sensing module. And personnel features, article features and environment features are extracted. The system also comprises an artificial intelligence analysis module and a user interaction module. According to the intelligent security dynamic identification management and control system, an artificial intelligence analysis module constructed by using a deep learning algorithm can perform real-time analysis and prediction on personnel behaviors, article states and abnormal events, discover potential safety hazards in advance and take corresponding management and control measures, thereby realizing conversion from passive defense to active prevention and control. For example, the system can give an alarm in time according to multi-dimensional information such as temperature change, smoke concentration and abnormal behaviors of people at the initial stage of a fire, so that precious time is won for people evacuation and fire extinguishment.

The invention relates to the technical field of three-dimensional scanning, and discloses a binocular three-dimensional scanning system and method with a texture mapping function. The system comprises a depth data acquisition module, a depth data correction module, a point cloud topology construction module, a texture mapping engine module and a mapping optimization module. The depth data acquisition module synchronously acquires a left view and a right view of a target object through a binocular camera array, and generates an original depth image data set; the depth data correction module performs abnormal value detection on the original depth data, extracts discrete distribution through aggregation boundary analysis and replaces abnormal values; the point cloud topology construction module generates a three-dimensional point cloud spatial topology structure, and associates point cloud coordinates with left view pixel positions; the texture mapping engine module projects left view RGB data to the point cloud to generate an initial texture map; and the mapping optimization module verifies the local color consistency through the generative adversarial network, and outputs an optimized three-dimensional texture model. The system improves the texture precision and authenticity of the three-dimensional model.

The invention relates to the technical field of artificial intelligence and computer vision, discloses a subway key component fault detection method and system based on an AI visual large model, and aims to solve the problems of low detection precision, weak generalization ability, insufficient multi-mode understanding, poor real-time performance and lack of state evolution modeling in the prior art. The method comprises the following steps: acquiring images of key components through a multi-view industrial camera array, and performing distortion correction, illumination normalization and noise suppression; a pre-trained visual large model is utilized to extract deep space features, and modeling is carried out on a continuous frame feature sequence through bidirectional LSTM to capture a time sequence change trend. By introducing the large-scale visual large model and spatio-temporal joint modeling, the identification capability of tiny defects is improved, the discrimination stability is enhanced, high-precision and low-delay automatic detection is realized, the false alarm rate and the omission ratio are remarkably reduced, and the detection efficiency and the system maintainability are improved.

The invention discloses a temperature dynamic control system for optical fiber optical wand processing. The temperature dynamic control system comprises an acquisition unit, a control unit, a learning unit and an integration unit. The acquisition unit captures production line material imaging data through a multi-macro camera array, acquires temperature, tension, humidity and speed data in real time by using a production line sensor network, and forms production state comprehensive data after edge computing node fusion preprocessing. The control unit establishes a dynamic mathematical prediction model based on the comprehensive data, predicts future temperature states of a plurality of temperature zones and generates an optimization control input instruction; the learning unit receives an optimization control input instruction and historical temperature data, and generates a self-adaptive control strategy by adopting a reinforcement learning algorithm; the integration unit integrates the control input instruction and the self-adaptive control strategy to form a final control instruction, the temperature of each temperature zone is adjusted, and the temperature control precision and the system stability are ensured.

The invention relates to the technical field of rock mechanical tests and safety monitoring, and discloses a rock fracture prediction method based on acoustic emission and surface strain cooperative monitoring, and the method comprises the steps: building a monitoring system comprising acoustic emission and an industrial camera array, and building a unified timestamp of each subsystem through a GPS time service module; through load triggering logic, a synchronous trigger is utilized to synchronously acquire an acoustic emission signal and a sample surface image sequence, and evolution characteristics of three-dimensional coordinates of an acoustic emission source and surface full-field strain data are solved. And on the basis, calculating a main strain field variation coefficient, and performing cross-correlation analysis on the resampled energy rate and strain rate. And finally, according to multi-parameter coupling criteria of acoustic emission time sequence parameters, positioning events and surface strain, identifying rock fracture precursor types, and predicting fracture moments and areas by combining a Voight model inverse velocity method and a seismic source projection technology.

The invention discloses sorting robot complete equipment based on an AI large model, the sorting robot complete equipment comprises a complete sorting assembly line, the assembly line is provided with an industrial camera array, a central control system, a mechanical arm execution module, a core calculation module based on the AI large model and the like, the industrial camera array collects video streams, and an encoder is in signal synchronization with an image collection module; the core calculation module converts video streams into information streams, the central control system makes decisions and plans grabbing tracks according to sorting category semantics and the information streams, a physical signal auxiliary evidence module and the like are further arranged, and all the modules work cooperatively. The technical effects that the positions, categories, materials and other information of the articles can be accurately recognized, efficient and accurate article sorting is achieved, the target category and the container distribution strategy can be adjusted in real time, and the motion parameters of the mechanical arm can be adjusted according to the material attributes of the articles are achieved.

The invention relates to a six-eye camera panoramic video real-time splicing method and system and computer equipment, and the method comprises the steps: synchronously carrying out the image collection through a six-eye camera array, and adding a timestamp to each video frame image; performing real-time feature extraction on the video frame image, and establishing a feature matching relationship in a view field overlapping region of adjacent cameras; based on the feature matching relationship, mapping each video frame image to a unified splicing coordinate system through projection transformation; performing splicing processing on the video frame images after projection transformation through multi-band fusion, and eliminating splicing traces at the boundary of the field of view; and reconstructing the panoramic frame images obtained after splicing processing according to a time sequence, and outputting a continuous panoramic video stream so as to realize real-time and seamless splicing and correction of six paths of high-definition video streams and finally output a global panoramic video stream with no visual fracture and consistent time and space.

The invention relates to a tunnel three-dimensional holographic splicing method and device for panoramic array camera networking, and the method comprises the steps: enabling a reference transmission camera array and a three-dimensional laser scanner on an inspection vehicle to move to different inspection positions along with the inspection vehicle for adjustment, achieving the transmission of peripheral control information at two ends of a tunnel into a measurement region, and constructing a dynamic control field; the limitation on measurement conditions is reduced; high-precision positioning and attitude determination are carried out based on fixed connection constraints of a look-around camera array and a three-dimensional laser scanner and a constructed dynamic control field, so that a measurement platform is unified to a global coordinate system, and the inherent defect of large accumulative error of a traditional three-dimensional modeling method is effectively avoided; and then the high-resolution look-around image is mapped to the three-dimensional point cloud data under the global coordinate system, so that high-resolution and high-precision three-dimensional holographic splicing of the tunnel ring surface under the global coordinate system is realized, global and local high-resolution and high-precision are achieved, and the acquisition precision of the tunnel three-dimensional hologram is greatly improved.

The invention provides a robot-based blade interior autonomous crawling sensing method and system, and the method comprises the steps: carrying a multi-sensor array on a robot, the multi-sensor array comprises a binocular fisheye camera array and a ToF sensor array, controlling the robot to start from a path travel starting point, obtaining a panoramic depth image based on the multi-sensor array, and transmitting the panoramic depth image to the robot; recognizing an obstacle based on the panoramic depth image, constructing a three-dimensional point cloud map based on the obstacle, avoiding the obstacle based on the three-dimensional point cloud map, generating a feasible path in the blade, and enabling the robot to move based on the feasible path until the robot reaches a path travel terminal point or a minimum space in which the robot cannot pass. Autonomous crawling perception of the robot in the blade can be ensured, movement of the robot in the blade is ensured, and detection is realized.

The invention relates to the technical field of circuit board detection, in particular to a quality detection system of an integrated circuit board, which comprises an image acquisition module used for cooperatively acquiring a surface image and a temperature distribution image of the integrated circuit board through a camera array and a multispectral light source; the preprocessing module is used for carrying out noise suppression, distortion correction and multispectral fusion on the surface image and the temperature distribution image and generating a standardized detection image; the defect detection module is used for extracting welding spot contour features, wire connectivity features and dielectric layer thickness features from the standardized detection image, the welding spot image and the inner layer wiring image of the integrated circuit board can be separated, interlayer interference is eliminated, multi-dimensional more accurate evaluation of defect risks is achieved, and the defect detection accuracy is improved. Misjudgment caused by traditional isolated analysis is avoided, the irradiation angle and intensity of the multispectral light source can be dynamically adjusted, and the reliability and efficiency of quality detection of the integrated circuit board are improved.

The invention provides a forging forming method and system for an aluminum alloy pipe target based on image processing, and relates to the technical field of image processing.The method comprises the steps that a three-dimensional digital grid model and a topological relation graph are established; a dual-spectrum camera array is adopted to obtain multi-view images and fuse the multi-view images; reconstructing a temperature radiation intensity mapping image through the neural radiation field network; executing image enhancement processing; extracting contour feature data by using a multi-scale texture decoupling network, and separating reflection, oxidation and forging components; a deformation trend is predicted based on the topological relation graph and the contour feature data, and a forging parameter adjusting instruction is generated; and the forging equipment is controlled according to the instruction and optimized in real time until the forming quality reaches the standard. According to the method, accurate monitoring and intelligent control of the aluminum alloy pipe target forging process are achieved.

The invention provides a panoramic image processing method and device suitable for an unmanned aerial vehicle and a panoramic unmanned aerial vehicle. The method comprises the following steps: acquiring multi-view images in real time through a multi-camera array of an unmanned aerial vehicle, acquiring pose data by using an inertial sensor, converting each view image into a first projection image and a second projection image, and performing dynamic target detection and recognition on the second projection image to obtain a dynamic target, geometric scene features and target distribution. Differentiated processing areas are divided, specifically, a high-precision algorithm is adopted for the first processing area, a low-precision simplified algorithm is adopted for the second processing area, and computing power focusing and processing capacity optimization are achieved. In the splicing stage, a dynamic target historical trajectory optimization splicing algorithm is combined, and processing strategies are adapted to different areas, so that the quality of the key areas is guaranteed, and the process of the non-key areas is simplified to reduce the calculation load. According to the scheme, the problem of real-time panoramic splicing under the condition that unmanned aerial vehicle hardware resources are limited is effectively solved, and the processing real-time performance and efficiency are remarkably improved.

The invention relates to an intelligent oil containment boom mechanical arm device and a control method thereof. The intelligent oil containment boom mechanical arm device comprises a mechanical arm, a hydraulic driving system, a sensing system and a control unit. The mechanical arm comprises a transmission base, a steering stand column, a large arm, a small arm, an overturning head and a tail end grabbing hook assembly. The hydraulic driving system comprises a hydraulic pump station, a proportional reversing electromagnetic valve group and a plurality of execution oil cylinders; the sensing system comprises a gyroscope array and a camera array to form a distributed attitude sensing network and a distributed visual sensing network; the control unit identifies and positions an oil containment boom target based on a sea surface image acquired by the distributed visual perception network, performs motion planning on the mechanical arm, and controls the mechanical arm to move and grab the target; in the target positioning process, posture changes of all parts of the mechanical arm are sensed in real time based on the distributed posture sensing network, and dynamic compensation is conducted on the posture error of the mechanical arm. According to the device, the oil containment boom can be quickly, accurately, reliably and intelligently grabbed, and the laying efficiency and safety of the oil containment boom are improved.

Provided are a camera-array-based intelligent measurement and control method and system for an ultra-high bridge tower (1). The system is provided with a camera array monitoring module, an image capture and processing module, a monitoring apparatus, a bridge tower (1) status evaluation and prediction module, and a bridge tower (1) intelligent control and early warning module. Targets (3) are vertically arranged on a bridge tower (1), and an array camera monitors the real-time displacement of each target (3) across the full height range of the ultra-high bridge tower (1); the monitoring apparatus acquires environmental parameters and stress data; a bridge tower (1) structure status model is established in the bridge tower (1) status evaluation and prediction module to predict the future alignment change trend of the bridge tower (1); and the bridge tower intelligent control and early warning module performs real-time early warning and control on the bridge tower (1). Thus, all-weather automatic monitoring and intelligent control of the alignment of the structure of a bridge tower (1) are realized, thereby effectively improving the measurement efficiency, reducing the labor costs, ensuring the accuracy and reliability of measurement results, and thus enhancing the construction quality of the bridge tower (1).

The invention relates to the technical field of ceramic intelligent manufacturing, in particular to a ceramic firing process thermal deformation intelligent prediction and self-adaptive temperature precise regulation and control system, which comprises a multi-source sensing module (a thermal infrared camera array, a temperature sensor, a laser displacement meter and a powder characteristic analyzer); a hybrid prediction model (a space-time convolutional neural network constrained by a physical mechanism); and the self-adaptive temperature control actuator comprises a partitioned burner group, a quenching / slow cooling pipeline and a gas-air volume decoupling controller. According to the scheme, through a multi-mode perception-hybrid modeling-dynamic optimization full-process closed-loop architecture, the single-point technical limitation of existing patents is broken through, and the zero-defect precise manufacturing target of ceramic firing is achieved.

The invention discloses a precise mark-free positioning system applied to an oral cavity. A camera array composed of a plurality of cameras which are annularly or specifically arranged is arranged on a mobile phone; an infrared imaging assembly for solving the problem of cooling water interference in the operation is designed on at least one camera; the processor is configured to integrate the acquired data including preoperative CBCT (cone beam computed tomography), oral scanning data and digital data of occlusal relationship to construct a complete preoperative oral three-dimensional model; establishing a spatial relationship between a visual system formed by registration of the turbine handpiece and the camera array and a preoperative oral cavity three-dimensional model, and obtaining adaptive relative poses among the parts; after the body position of a patient is fixed during an operation, a complete oral environment point cloud is formed for each frame of input data currently collected by the visual system, and the complete oral environment point cloud is fitted with a patient oral model obtained through oral scanning and CT scanning in advance, so that the oral position of the patient is positioned in real time.

The invention provides a high-precision multi-view image recognition detection method based on a CCD camera, and the method comprises the steps: employing a convolutional neural network to extract the noise perception features of each view image through a multi-view image set shot by a CCD camera array; carrying out semantic segmentation by adopting a self-adaptive multi-view semantic segmentation network; constructing final feature representation of the target area based on the enhanced features and the key features; performing view angle consistency matching based on the final feature representation of the target region by adopting similarity calculation; and based on the final target feature map, combining a classification module to identify the defect type, and outputting the type and position information of the defective product. According to the method, the noise removal of the multi-view image is realized through a self-supervised learning method, and the image features under different views are optimized in combination with a self-adaptive semantic segmentation technology, so that the accurate target recognition of the multi-view image is realized.

The invention provides a tunnel point position change automatic observation method and system combined with an AI algorithm, and relates to the technical field of tunnel monitoring, and the method comprises the steps: employing a speckle projector to form an irregular speckle pattern on a tunnel wall surface, and carrying out the multi-angle collection through a polarized light camera array, and obtaining the speckle intensity data; pre-embedding a coding microstructure generated based on quantum state superposition in the speckle pattern; establishing a speckle-stress correlation analysis model according to the speckle intensity data, and calculating three-dimensional displacement data in combination with the deformation data; and inputting the environmental monitoring data, the stress distribution data and the displacement data into the self-evolution heterogeneous feature network for analysis and prediction. According to the method, the measurement precision is improved by adopting the quantum coding speckles, the dynamic correlation analysis of multi-source data is realized through the self-evolution heterogeneous feature network, and the environmental adaptability and prediction accuracy of the system are enhanced.

The invention relates to the technical field of industrial visual inspection, in particular to a cigarette packet surface defect detection method and system and a storage medium, and the method comprises the following steps: optimizing a light field by adopting a multi-mode light field regulation and control model; a distributed camera array is arranged, and a multi-angle image, a multi-mode image and a multi-polarization-angle multi-spectral image of the surface of a to-be-detected cigarette packet are obtained; according to the multi-angle image and the multi-modal image, obtaining a three-dimensional reconstruction depth map by adopting a depth estimation model; based on the multi-polarization-angle multispectral image, defect detection is carried out, and a defect edge is extracted; constructing a defect classification model based on self-supervised learning; carrying out joint training on the multi-modal light field regulation and control model, the depth estimation model and the defect detection and defect classification model to obtain a joint detection model; and performing defect detection on a to-be-detected cigarette packet by adopting the trained joint detection model. According to the embodiment of the invention, by fusing the dynamic adjustment light field, the deep learning algorithm and the multi-angle imaging, the defect detection performance in the high-reflection environment is optimized.

A ship sub-assembly weld seam identification method, comprising: creating a point cloud weld seam identification training dataset by means of point cloud discretization, virtual sampling, weld-zone semi-automatic interactive annotation, and style transfer-based augmentation of ship sub-assembly CAD data; training a point cloud weld seam identification model, and exporting a weld seam identification model data file; using a 3D point cloud camera array as a sampling camera to collect workpiece 3D point cloud data of a sub-assembly workpiece to be welded at the production site; by means of the weld seam identification model data file, obtaining weld seam annotation position data and positioning a weld seam; and issuing a welding process plan and instruction to control a welding robotic arm and a control device therefor to complete sub-assembly workpiece welding. Also provided are a ship sub-assembly weld seam identification system, a computer device, and a computer-readable storage medium. The method overcomes the shortcomings in conventional 3D weld seam identification methods such as point cloud incompleteness and identification failures caused by steel types and the imaging angles of 3D cameras, thereby greatly improving the adaptability of automatic weld seam identification methods to complex workpieces.