109 results about "Visual defects" patented technology

The invention discloses a printed circuit board withstand voltage detection and evaluation method, which comprises the following steps: in the whole withstand voltage detection process, aiming at different process batches and environments, acquiring electrical signals, process parameters, environment variables and multi-modal visual data, and carrying out time sequence synchronization marking and standardized preprocessing; extracting multi-source features by using a feature coding algorithm, and establishing a multi-mode distribution model covering signal mutation, visual defects and working condition anomalies; through cross-modal feature fusion and a time sequence causal inference model, an abnormal event causal relationship graph is automatically constructed, and a voltage withstanding failure root cause chain is identified and positioned; according to the method, the efficient structured management of the detection data, the accurate traceability of the abnormal reasons and the cross-batch analysis are realized, and the capability of diagnosing and preventing the withstand voltage abnormality of the printed circuit board is remarkably improved.

The invention discloses a few-sample visual defect detection method based on approximate proximity search, and belongs to the technical field of industrial intellectualization. The system mainly comprises an image acquisition module, a feature extraction module, a memory bank construction module, a feature compression module, an approximate nearest neighbor search module and an anomaly detection and result output module. The feature extraction module is used for performing feature extraction on the collected image; the feature compression module is used for reducing the feature data volume and reducing the internal memory and calculation burden; the memory bank construction module is mainly operated during system initialization or model training and is used for constructing a feature memory bank of normal samples; the approximate nearest neighbor search module is responsible for quickly finding normal sample features closest to the input features in the memory bank; and the anomaly detection and result output module is used for converting distance information obtained by neighbor search into anomaly. According to the few-sample visual defect detection method based on the approximate proximity search, the effect of few-sample anomaly detection is achieved.

The invention provides a roadway surface and internal defect synchronous sensing method based on point cloud registration. The roadway surface and internal defect synchronous sensing method comprises the following steps: S1, multi-sensor joint calibration; s2, synchronously acquiring data; s3, data preprocessing; s4, projecting a camera image to the point cloud; s5, surface defect identification: S5.1, geometric defect identification based on point cloud; s5.2, performing visual defect identification based on the image, and marking an identification result in the point cloud; s6, integrating the geometric defect point cloud and the visual defect point cloud; s7, identifying internal defects; s8, determining an internal confirmation three-dimensional position; s9, marking an internal defect point cloud; and S10, uniformly visualizing the surface defects and the internal defects. According to the invention, through cooperative work of the laser radar, the ground penetrating radar and the camera, the same perception of tunnel defect surface and internal defects is realized through point cloud visualization.

The invention provides an energy storage device damage diagnosis and interaction system based on penetration vision and a large model, and relates to the technical field of energy storage device damage diagnos.The energy storage device damage diagnosis and interaction system comprises a penetration type 3D sensing module, a periodic topology analysis module, a feature mapping and retrieval module and a large model reasoning and interaction module, and the energy storage device is scanned and reconstructed; the method comprises the following steps: constructing a time-space decoupling periodic Transform network, introducing a periodic mask matrix to force the network to pay attention to a repeatability rule of an internal structure of a battery, and identifying internal tiny deformation and structural damage by calculating topological consistency under the condition that a large amount of negative sample training is not needed; visual defect features are mapped into text embedding by utilizing a feature projection technology, a diagnostic report containing physical cause analysis and maintenance suggestions is generated by combining a retrieval enhancement generation technology and a large language model, and a user is supported to perform interactive questions and answers in a natural language. The method can solve the problems that in the prior art, three-dimensional deformation is difficult to quantify, small samples are difficult to train, and intelligent decision-making ability is lacked.

The invention relates to a training method of a defect detection model and a PCB defect detection method, and the training method comprises the steps: carrying out the feature extraction based on a first training image through a first network branch of an initial detection model, and obtaining a visual defect feature map; performing feature extraction based on a second training image through a second network branch of the initial detection model to obtain a thermodynamic defect feature map; fusing the visual defect feature map and the thermal defect feature map by using the initial detection model to obtain a composite defect feature map, and performing detection based on the composite defect feature map to obtain a prediction result; and constructing a physical constraint item according to the space gradient information of the thermodynamic defect feature map, constructing a joint loss function according to the defect label, the prediction result and the physical constraint item, and performing iterative training on the initial detection model based on the joint loss function to generate a defect detection model. According to the invention, the problem of low accuracy of PCB defect detection is solved.

The invention discloses a multi-station synchronous plastic cup spoon full-surface visual defect detection device and method, and relates to the technical field of visual detection.The multi-station synchronous plastic cup spoon full-surface visual defect detection device comprises a multi-station rotating platform, multiple sets of flexible jigs used for clamping parts to be detected, a multi-station detection module and a detection controller; the four detection modules sequentially cover surface texture, three-dimensional morphology, material characteristics and mechanical property full-dimensional defects of a to-be-detected part; 2D characteristics such as surface smudginess, scratches and printing defects are captured through a multispectral camera; the 3D point cloud reconstruction module identifies three-dimensional shape defects such as pits and deformation through laser or structured light scanning; the hyperspectral material detection module is used for identifying material variation and wear degree through spectral analysis; the dynamic stress test module detects hidden mechanical defects such as stress whitening and microcracks through a loading test.

The invention relates to the technical field of image acquisition, in particular to a non-contact visual defect acquisition system and defect detection method for disc scissors, comprising an image acquisition device and a control device electrically connected with the image acquisition device; the image acquisition device comprises a camera and an LED stroboscopic lamp, and by combining the stepping distance between acquisition frames and the acquisition position of the camera with the rotation speed of the disc shear tool, the image acquisition rhythm matched with the high-speed rotation of the disc shear tool can be realized, the detailed information of each part of the disc shear tool can be quickly acquired, and the image acquisition efficiency is improved; through matching of the PWM stroboscopic control frequency of the LED stroboscopic lamp and the exposure time of the camera, it is ensured that stable and appropriate illumination intensity and illumination time are provided when the camera is exposed every time, the problems of image blurring, inconsistent brightness and the like caused by uneven illumination or illumination change are solved, and the system is simple in structure, convenient to operate and low in cost. The problems that a traditional disc type cutter is low in detection efficiency, large in error and lagged in detection are effectively solved.

The present application belongs to the field of machine vision defect detection, and in particular to a thermocell electrode sheet vision defect detection method and a vision defect detection device. Product size information of a thermocell electrode sheet is obtained, and the field of view of a camera is configured. A thermocell electrode sheet to be detected is placed on a detection platform, front surface images, side surface images, back surface images, and inner hole images of the thermocell electrode sheet are captured by multiple direction detection cameras, and are processed to obtain the to-be-detected images of the front surface images, the side surface images, the back surface images, and the inner hole images. According to the to-be-detected images of the front surface images, the side surface images, the back surface images, and the inner hole images, a trained YOLO8 recognition model is used to describe and determine defects.

The invention discloses an AI visual defect identification and detection method integrated with a light supplementing module, and relates to the technical field of computer vision, and the method comprises the following steps: S1, data acquisition: carrying out multi-angle shooting on a product at a fixed station of a production line by using an industrial camera integrated with the light supplementing module, the light supplementing module automatically adjusts the illumination intensity and angle according to product surface characteristics and ambient light conditions; s2, image preprocessing; s3, feature extraction; s4, defect classification / positioning; and S5, outputting a result. According to the AI visual defect identification and detection method integrated with the light supplementing module, the light supplementing module is integrated, and the programmable LED light source array is adopted, so that the change of the illumination intensity of a production line can be sensed in real time, and the illumination intensity and the angle can be automatically adjusted, and therefore, the problem that the local part of an image is too bright or too dark due to ambient light fluctuation can be effectively avoided; especially for highly reflective materials such as metal and coated surfaces, the light supplementing module can select the angle combination with the weakest reflected light for light supplementing.

The invention relates to the technical field of computer vision, and discloses an industrial product visual defect detection method and system based on multi-task reverse knowledge distillation, by introducing a multi-scale projection layer and a multi-task learning framework, the expression ability of a model to normal sample features and the inhibition ability to abnormal signals are significantly improved, and the detection accuracy is improved. Therefore, industry-leading defect detection and positioning precision is realized on industrial reference data sets such as MVTec, extremely low reasoning delay and memory occupancy are kept, the reasoning speed is far higher than that of mainstream methods such as PatchCore, an efficient and accurate technical scheme is provided for industrial large-scale real-time defect detection, low delay and low memory consumption are kept, and the defect detection accuracy is greatly improved. The abnormal detection and positioning precision of industrial product visual defect detection is remarkably improved, and the defect that performance and efficiency are difficult to consider in the prior art is overcome.

The invention relates to the technical field of visual inspection, and provides a lens flaw detection method based on deep learning and related equipment thereof. Visual defect feature extraction is performed on an original lens image to obtain a defect candidate region set, multiple feature discrimination is performed on the defect candidate region set in combination with a visual algorithm to obtain an initial defect region set, and an area threshold and the initial defect region set are compared and screened to obtain a to-be-detected region set. And performing forward reasoning on the to-be-detected region set through a ShuffleNetV2 depth model to obtain a defect type and a confidence value of each region, performing defect judgment on the defect type of each region according to a confidence threshold and the confidence value to obtain lens defect data, and performing defect elimination according to a defect elimination mode and the lens defect data. Through visual preprocessing, multi-feature discrimination, region screening and lightweight deep model reasoning, high-precision, low-delay, easy-to-deploy and high-robustness lens flaw detection is realized.

The invention discloses a visual defect detection method, device and equipment based on knowledge prompt and a storage medium, and relates to the technical field of appearance defect detection.The method comprises the steps that multi-dimensional image data corresponding to a to-be-detected object is collected, and the multi-dimensional image data comprises a multi-band image, a multi-polarization image and an active structure light phase image; generating knowledge prompt information of the to-be-detected object according to a preset knowledge graph; and performing fusion defect detection based on the multi-dimensional image data and the knowledge prompt information to obtain a defect detection result. According to the method, the comprehensive information of the to-be-detected object is obtained from different dimensions by collecting the multi-dimensional image data, semantic guidance is provided by combining the knowledge prompt information generated by the knowledge graph, cross-modal fusion defect detection of the to-be-detected object is achieved, and therefore the visual defect detection precision is improved.

This invention discloses a machine vision-based method and system for identifying defects in wire ropes of port cranes, belonging to the field of visual defect recognition technology. The method includes: real-time acquisition of wire rope images; real-time acquisition of the wire rope's motion state data, including acceleration and angular velocity, via an inertial measurement unit; prediction of the wire rope's current motion state using a Kalman filter algorithm based on the previously acquired motion state data; construction of a fuzzy kernel function based on the predicted current motion state, and sharpening of the currently acquired wire rope image using an inverse filtering algorithm; and defect identification using a target detection model on the sharpened image. This invention enables rapid and accurate detection of surface defects in wire ropes under dynamic fuzzy and complex environments.

The invention discloses a visual defect identification and evaluation method for aviation I-shaped edge strip processing quality, particularly relates to the field of machine vision detection and discrimination, and is used for solving the problem of mistaken elimination or length measurement distortion caused by easy confusion of end part short region slender defects and end surface mirror reflection secondary images. The method comprises the following steps: establishing part local coordinates and fillet belt unfolding coordinates; screening a main evaluation view angle and a dual view angle and generating a displacement prediction window; extracting a candidate slender response set through dual-view collection and local registration; calculating a mapping residual distance and a same shift segment length and generating a mirror pseudo coefficient; and executing gating judgment and mirror image mask removal or length measurement and category judgment, and binding and storing to finish evaluation.

The invention relates to the technical field of visual defect detection, in particular to an intelligent dam surface defect detection method based on computer vision, which comprises the following steps: S1, acquiring a dam surface image sequence to form an original detection image set; s2, performing adaptive light field compensation and noise suppression processing on the original detection image set; s3, multi-scale region segmentation is carried out, curvature features are extracted, and a structural feature map is generated; s4, executing three-dimensional texture reconstruction on the basis of the structural feature map, and extracting a surface topography feature vector; s5, defect feature coupling analysis is carried out, and a defect classification matrix is generated; and S6, performing coordinate mapping in combination with a geographic information system, and outputting a defect detection report containing space coordinate information. According to the method, through fusion of multispectral imaging, three-dimensional texture reconstruction and geographic coordinate mapping technologies, high-precision identification and spatial positioning of dam surface defects are realized, and the detection accuracy and the operation and maintenance response efficiency are remarkably improved.

The application belongs to the technical field of optical lens detection, and particularly relates to an optical lens appearance defect detection device, which comprises a supporting assembly, a feeding device is installed on the supporting assembly, a first discharging slide plate is installed at the discharging end of the feeding device, a mounting seat is fixedly connected to the supporting assembly below the first discharging slide plate, supporting rollers are rotationally connected to both ends of the mounting seat, and the cooperation of the structures such as the belt ring, the first visual defect detection camera, the pushing cylinder, the guide plate and the driving roller is achieved, so that after the optical lens is accurately positioned in the positioning groove formed in the belt ring, the first visual defect detection camera arranged on both sides is used to detect the end of the optical lens, then the pushing cylinder is used to push the optical lens to the driving roller, the driving roller is rotated to drive the optical lens to slowly rotate, the second visual defect detection camera is used to detect the circumferential side of the optical lens, and the mechanical hand is not needed to be used for carrying and aligning, so that the detection efficiency is greatly improved.

The application relates to the technical field of machine vision and image processing, and discloses a plastic valve visual defect real-time detection method and system based on deep learning, which comprises the following steps: acquiring a gray image containing a plastic valve, positioning the centroid of a plastic valve main body area by using geometric moments; converting the gray image to a polar coordinate based on the centroid, applying one-dimensional high-pass filtering, and determining a dynamic segmentation threshold to extract abnormal gray pixel points; performing spatial density clustering after reverse mapping to determine a suspected defect area; calculating the Euclidean distance between the centroid of the suspected defect area and the centroid of the plastic valve main body area, synchronously inputting the distance and image features into a double-input convolutional neural network to output a plastic valve defect detection result. The application filters out cloud and fog-shaped diffuse reflection interference, and improves the accuracy of defect detection.

The invention relates to a visual aid (1) comprising at least one imaging optic (2) with variable focal length for positioning between a viewer's eye and a scene to be viewed, further comprising means for controlling the change in focal length of the at least one imaging optic, optoelectronic means (3) for determining the distance of the imaging optic (2) optionally from one or more nearest objects in the scene to be viewed in the viewing direction of the viewer's eye and / or with a selectable angular tolerance around the viewing direction of the viewer's eye, electronic means for controlling the means for controlling the change in focal length of the at least one imaging optic (2), wherein said determined distance is converted into a target focal length of the at least one imaging optic (2) by means of a table of values ​​and / or by means of an equation.as well as an energy source for the aforementioned optoelectronic and electronic means and the means for control. The invention further relates to a method for vision correction and a computer program.

The invention relates to the field of wafer machine vision defect classification, and discloses a wafer machine vision defect classification device which comprises a bottom plate, a supporting column is arranged in the middle of the upper end of the bottom plate, a telescopic rod is arranged at the upper end of the supporting column, and a first rotating rod is connected to the upper end of the telescopic rod. A first inner rotating shaft is connected to the overlapping position of the first rotating rod and the telescopic rod, a second rotating rod is arranged at the upper end of the first rotating rod, a second inner rotating shaft is connected to the overlapping position of the second rotating rod and the first rotating rod, and a third rotating rod is arranged at the upper end of the second rotating rod. The wafer center is accurately positioned, it is ensured that the geometric center of the wafer is aligned with a visual system during detection, the position error is eliminated, the defect classification precision is improved, the supporting round rod serves as a transfer platform, a temporary placement point is provided for the wafer, the wafer center is preliminarily centered, and the subsequent positioning step is simplified; and accurate positioning of the circle center of the wafer is realized through flexible pressing.

The application discloses a kind of based on improved AnoGAN small sample substation equipment visual defect detection method, belong to machine learning technical field;The method comprises: obtaining training substation equipment image;According to normal substation equipment image training AnoGAN adversarial generation network;Defect substation equipment image is input into AnoGAN adversarial generation network after training is completed;After pre-processing, defect substation equipment image is input into pre-training similarity classification network and is classified;After pre-processing, defect substation equipment image is input into pre-training convolution classification network and is classified;According to similarity classification result and convolution classification result, obtain defect detection result.The application also provides a kind of based on improved AnoGAN small sample substation equipment visual defect detection system.The network of the application is trained to normal data for training, so its generation effect can be distinguished to all non-normal state equipment, and the practical application is more widely used.

The application discloses a kind of for carbon fiber precursor's defect detection device, it is related to carbon fiber quality detection equipment field, including defect detection mechanism, the inside of defect detection mechanism is equipped with precursor amplification mechanism, carbon fiber precursor is inserted in the inside of precursor amplification mechanism, the back of precursor amplification mechanism is equipped with rack fixing seat B, first convex mirror is provided on precursor amplification mechanism, second convex mirror is installed in the side of first convex mirror, third convex mirror is provided in the side of second convex mirror away from first convex mirror, fourth convex mirror is installed in the side of third convex mirror away from second convex mirror;Visual defect detector is provided on defect detection mechanism.The used for carbon fiber precursor's defect detection device, precursor amplification mechanism can amplify detection surface, so that visual defect detector can more clearly observe the subtle features and potential defects on the surface of carbon fiber precursor, improve production efficiency, by multidirectional detection to precursor, improve detection accuracy.

The application discloses a kind of wind power blade defect detection method and system, it is related to defect detection technical field, the system includes: external data acquisition module, state evaluation module of lightning arrester, inner cavity check mapping module, inner cavity image acquisition module, inner cavity defect evaluation module, blade health evaluation module.The application generates internal key check area instruction based on external defect coordinates, guides inner cavity detection equipment to carry out fixed-point review, improves the pertinence and efficiency of inner cavity detection;At the same time, by fusing lightning arrester electrical performance data and visual defect information, and the defects in space are associated with correction, the specificity of lightning damage is realized, the accuracy and reliability of wind power blade health state comprehensive evaluation are improved, the problems that the internal detection blindness is large, the efficiency is low in the prior art, and the defect cause judgment is single, and lightning damage and ordinary defects cannot be accurately distinguished.

This invention belongs to the field of machine vision inspection, specifically relating to a deep learning-based visual defect detection system for injection-molded shells of high-end equipment. The system includes: a biomimetic compound-eye programmable optical imaging subsystem for generating a reverse compensation light field based on surface morphology to achieve feature enhancement and noise stripping; a production process perception module for monitoring environmental parameters and extracting state feature vectors; a cloud-based adaptive learning center for constructing a meta-learning model and generating specific adaptation parameters; and an edge-side multi-task inference terminal for receiving dynamic knowledge distillation results and performing defect localization and pixel-level segmentation. This invention, through physical imaging and algorithmic collaboration and a cloud-edge architecture, suppresses optical overflow and solves the model adaptation problem under production fluctuations, improving processing efficiency while maintaining accuracy, and achieving closed-loop optimization and continuous evolution of the system.

The invention relates to a method for manufacturing a customized optical element to adjust an optical property of an optical component, such as goggles, wherein the method comprises the steps of receiving customer data indicative of a visual defect a person and / or a desired visual correction of the optical component; determining, from the customer data, the shape of a first surface of an optical element to be manufactured, such that the optical component, when the optical element is applied to an optical surface of the optical component, compensates at least partially or entirely the visual defect of the person or exhibits the desired visual correction; manufacturing the optical element using a liquid material that when brought in shape, is solidified.

The utility model relates to the technical field of visual detection, in particular to visual defect detection equipment which comprises a workbench, a protective cover covering the workbench, a backlight plate fixed on the workbench, a clamp installed on the workbench and a visual camera assembly arranged above the workbench. The detection driving assembly is used for driving the visual camera assembly to move left and right, the protective cover is installed on the workbench, the detection driving assembly is installed on the inner side of the top of the protective cover, the clamp is used for clamping and positioning a product on a backlight plate, and a controller is arranged on the protective cover; the device can detect the defects of uneven density, holes, scratches, smudginess, pits and the like on a product, avoids the situation that the defects of the holes, the pits and the like are not observed, and ensures the quality of the product.

This utility model relates to the field of automotive seat processing technology, specifically to an automotive seat visual defect detection device. It includes a base, a servo motor that outputs controllable torque to drive the entire moving assembly to perform detection movements, and a camera that collects image data of the automotive seat surface for subsequent defect analysis. This automotive seat visual defect detection device smoothly transports the automotive seat to the detection area below the protective cover via a conveyor belt. The servo motor drives a gear and a gear ring to precisely mesh, causing the camera to reciprocate in a circular motion along a guide ring. This innovative motion method overcomes the blind spot problem inherent in traditional straight-line detection. The combined design of the connecting frame and damper effectively suppresses the impact of mechanical vibration on image quality. The annular protective cover, combined with the lower slot structure, protects the internal components while ensuring an open detection field of view. Compared to fixed-point detection schemes, this design improves detection coverage.

The utility model provides a turnover type AOI visual defect double-sided detection device. The turnover type AOI visual defect double-sided detection device comprises a bottom plate; the supporting frame is fixedly connected to the top of the bottom plate, a detection table is fixedly connected to the top of the supporting frame, the detection table is fixedly connected to the mounting frame, a through detection opening is formed in one side of the mounting frame, and sliding rings are fixedly connected to the positions, close to the two sides, of the inner surface of the detection opening. And the inner surfaces of the two sliding rings are rotationally connected with a rotating frame through two sets of pulleys, photographing openings are formed in the positions, located on the front face and the back face, of the outer surface of the rotating frame, and a driven gear is installed at the position, close to one side, of the outer surface of the rotating frame. By means of the design, in the AOI double-face detection process of the plates, the plates can be turned over automatically, the plates are detected in the conveying process, and the AOI double-face detection efficiency of the plates is greatly improved.

The invention relates to the technical field of visual defect detection, in particular to a visual defect detection method of a release agent for film adhesive production, which comprises the following steps: mixing solvent samples under different sampling conditions with a fluorescent agent, and sequentially coating in a plurality of local areas of film adhesive; irradiating by utilizing an excitation light source, and judging whether the phase distribution state of the solvent sample is different or not based on a characteristic quantitative comparison result of a brightness space distribution curve; according to a numerical value comparison condition in the brightness value sequence, determining a clustering feature category; and finally, determining the qualification of the release agent solvent based on curve information, or determining the distribution position of the defective release agent solvent based on a numerical fluctuation analysis result in the brightness value sequence. Furthermore, the solvent phase distribution state of the release agent is directly quantified and compared through visual inspection, the defect area is accurately positioned, the quality defect positioning accuracy of the release agent is improved, and the production quality stability of the adhesive film paper is improved.

The invention belongs to the technical field of optical lens detection, and particularly relates to an optical lens appearance defect detection device which comprises a supporting assembly, a feeding device is installed on the supporting assembly, a first discharging sliding plate is installed at the discharging end of the feeding device, and a mounting base is fixedly connected to the portion, below the first discharging sliding plate, of the supporting assembly. The belt ring, the first visual defect detection camera, the material pushing air cylinder, the material guiding plate, the driving roller and other structures are arranged in a matched mode, so that the optical lens is accurately positioned in a positioning groove formed in the belt ring; the first visual defect detection cameras arranged on the two sides are used for carrying out defect detection on the end portion of the optical lens, then the pushing air cylinder is used for pushing the optical lens to the driving roller, the driving roller rotates to drive the optical lens to rotate slowly, and the peripheral side of the optical lens is detected in cooperation with the second visual defect detection cameras. A manipulator is not needed for carrying and aligning, so that the detection efficiency is greatly improved.

The invention provides a method and a system for quickly establishing a visual defect library. The method comprises the following steps: irradiating a sample by using a first visible light beam and marking the sample as an initial marking area; on the basis of the initial labeling area, irradiating by using a second visible light beam and removing an irradiated area of the second visible light beam from the initial labeling area in real time to obtain an ROI labeling result; adjusting light source parameters and image acquisition angles, repeating the irradiation marking step, and acquiring enhanced sample images under different light source parameters and different angles; and a training data set is constructed based on the sample image and is used for training an image detection model. According to the method, the ROI region of the sample is pre-labeled in the data acquisition stage through the visible light beam device, so that the data labeling cost and the post-processing difficulty are effectively reduced, the deviation problem of automatic labeling of a diffusion model is avoided, and efficient and reliable rapid establishment of a visual defect library is realized.