47results about How to "Improve defect detection rate" patented technology

The invention provides a battery film production automation control method and system based on visual inspection, and relates to the technical field of production control, and the method comprises the steps: collecting a film image, carrying out the fusion, generating a fused film image, carrying out the noise reduction, constructing a feature point matching relation, removing wrong point pairs, carrying out the geometric correction, carrying out the sub-pixel-level alignment, and obtaining a registration image; edge enhancement is carried out based on an improved Frangian filtering algorithm, an edge region is repaired and segmented to obtain a film edge contour, three-dimensional reconstruction is carried out, three-dimensional point cloud coordinates are solved, a film three-dimensional contour model is reconstructed, a film defect detection model is constructed, and a film defect distribution diagram and defect confidence are obtained through forward propagation; performing association analysis with a production process, constructing a feature mapping model, generating a defect knowledge base and matching a process parameter optimization strategy, generating a dynamic instruction and sending the dynamic instruction to a station control unit, monitoring operation data, performing closed-loop feedback control, and obtaining a current optimal control scheme.

The invention discloses an array eddy current C-scan image enhancement method based on image pyramid reconstruction, and the method carries out the combined bilateral filtering of an array eddy current C-scan original image and a C-scan image after Gaussian filtering, and can remove most noisy points while keeping the detail information of small defects. According to the method, an image pyramid algorithm is adopted to carry out Laplacian pyramid decomposition, detail images under different scales are obtained, detail information of the images is further extracted, and noise is effectively removed. According to the method, the detail images under different scales are subjected to linear superposition reconstruction according to the proportion of L2: L1: L0 being 1: 0.2: 0.1, and detail information under multiple scales can be combined, so that the defect detection rate is improved to the maximum extent, the signal-to-noise ratio is further improved, and the array eddy current C-scan image with the high signal-to-noise ratio is obtained.

The invention belongs to the technical field of weld joint detection, and particularly relates to a weld joint defect detection method and device. The method comprises the following steps: S1, generating a magnetic field at a weld joint to be detected by using an excitation probe; s2, original magnetic field signals of the magnetic field in the X axial direction, the Y axial direction and the Z axial direction are collected; s3, processing the original magnetic field signal in each axial direction by using wavelet transform to obtain characteristic magnetic field signals Bx, By and Bz containing defect characteristics in three axial directions; s4, respectively calculating amplitude difference values between wave crests and wave troughs in the oscillogram of the characteristic magnetic field signals Bx, By and Bz; and if at least one of the amplitude difference values in the three oscillograms is greater than a preset threshold value, determining that the weld joint has defects. Through the excitation probe and a magnetic field signal processing method matched with the excitation probe, noise interference can be removed to accurately detect defects and estimate reference sizes of the defects, so that the purpose of improving the detection accuracy is achieved.

The utility model relates to the technical field of pipe detection, in particular to an enhanced design of detection precision of a detection device. The utility model discloses a steel pipe surface defect detection device which can keep high-precision dynamic stability in the detection process, provide a clean detection environment and optimize the imaging precision of the steel pipe surface defect detection device. The detection device comprises a shell, an imaging assembly arranged in the shell and a clamping assembly used for clamping a steel pipe, and the shell is provided with a dustproof ring used for allowing the steel pipe to pass through. The clamping assembly comprises a downward pressing device moving in the vertical direction and a supporting roller arranged on a downward pressing path of the downward pressing device, and the supporting roller is used for supporting the bottom of the steel pipe and guiding the steel pipe to enter the shell through rolling movement.

This invention relates to the field of intelligent sensing and detection technology, and provides a quality inspection device for cardboard box printing, including an adsorption module and an intelligent sensing quality inspection module. The intelligent sensing quality inspection module includes a detection industrial control computer, which internally includes a multispectral image acquisition unit, a deformation compensation unit, a multi-dimensional feature extraction unit, and an adaptive stitching unit. The adaptive stitching unit includes a multi-pose fusion subunit, configured to fuse images acquired and processed under at least two different adsorption poses, and utilize the complementary information of the occluded area under different poses to generate an unobstructed full-page inspection image. This device, by setting a positioning frame with a hollow plate, uses negative pressure adsorption to flatten each side panel of the cardboard box to the same horizontal reference plane, effectively solving the problem of bending and inconsistent imaging of adjacent side panels due to residual stress from creases after the cardboard box is unfolded, and providing a unified imaging benchmark for subsequent inspection.

The utility model discloses a titanium alloy pipe sealing detection equipment relates to industrial detection and quality control technical field, including bottom plate, moving assembly, sealed connection subassembly, nitrogen storage and delivery subassembly and detection component, the bottom plate upper end is equipped with moving assembly, nitrogen storage and delivery subassembly and detection component, moving assembly contains dovetail type sliding slot, two -way threaded rod no.

The utility model relates to the field of weld ray detection, disclose a kind of weld detection is with the fixed device of radioactive source, the fixed device includes base and the support rod for installing the source guide tube of gamma-ray source connected on base, the base includes a adjustable level base plane, the support rod includes upper support rod and lower support rod, lower support rod is slidable connection relative to base, upper support rod is slidable fixed connection relative to lower support rod, the sliding direction of lower support rod, upper support rod is always perpendicular to the base plane, upper support rod is used as source guide tube support for fixing source guide tube parallel to the base plane. When using, first level base plane can be adjusted outside container and according to the radial pre-positioning of container inner diameter to fixed device, axial positioning is completed inside container, the fixation of source guide tube is reliable, positioning accuracy is high, and it is convenient to operate, more safe, easily guarantee detection quality.

The present disclosure relates to a mask defect detection system, comprising: a vacuum cavity and a vibration isolation platform, wherein the vacuum cavity is located above the vibration isolation platform, and further comprising: a laser used as a system light source to generate laser light; a beam shaping system used to perform beam shaping on the laser light; a decoherence mirror used to achieve decoherence of the laser light; a pupil conversion wheel used for pupil selection; a zone plate used to image defects on a mask, a flange used to achieve a sealed connection between cavities; a four-claw slit used to fine-tune the size of an illumination spot; the decoherence mirror, the pupil conversion wheel and the zone plate are located in the vacuum cavity; the laser light generated by the laser passes through the beam shaping system, the flange and the four-claw slit, the decoherence mirror and the pupil conversion wheel in sequence, and is irradiated onto a mask for mask defect detection.

This application discloses a method, equipment, medium, and product for detecting defects in fillet welds of B-type sleeves in oil and gas pipelines, relating to the field of non-destructive testing. The method includes constructing a simulation model and a phased array ultrasonic probe model in simulation software, and performing sound beam propagation simulation and defect response simulation on the phased array ultrasonic probe model and the simulation model to obtain the optimal placement position and optimal focusing law parameters of the phased array ultrasonic probe. The simulation model includes a simulation model of the workpiece under inspection and preset defects on the fillet weld. The simulation model of the workpiece under inspection includes: a B-type sleeve, the fillet weld, and the main pipeline. The B-type sleeve is welded to the main pipeline via the fillet weld. Based on the optimal placement position and optimal focusing law parameters of the phased array ultrasonic probe, the fillet weld of the workpiece under inspection is scanned using phased array ultrasonic scanning to achieve fillet weld defect detection. This application can improve the reliability of the detection results.

This invention discloses a wafer surface defect detection system and method based on a high-speed line scan camera, belonging to the field of wafer defect detection technology. It includes: an illumination system for providing adjustable bright-field illumination and illumination angle, and forming adjustable-size line spots on the wafer surface through a line spot generation device; a wafer scanning assembly including a stage for supporting the wafer and a displacement control assembly for driving the stage movement; an imaging system including a bright-field imaging assembly; a high-speed line scan camera for high-speed scanning and imaging of the signals collected by the imaging system, and balancing the signal-to-noise ratio, dynamic range, and detection resolution through a programmable pixel merging function; and a control system for coordinating and controlling the various system components and achieving parameter optimization and online adaptive calibration. This invention can achieve high-resolution, high-sensitivity wafer surface micro / nano defect detection while ensuring high detection speed.

This invention provides a defect detection method, a model training method, an apparatus, a device, a medium, and a product. The detection method includes: acquiring an image of a battery cell sealing ring to be inspected, and a defect-free front image and a defect-free back image of a sample battery cell sealing ring; constructing data from the foreground target pixels of the image of the battery cell sealing ring to be inspected, the defect-free front image, and the defect-free back image of the sample battery cell sealing ring to obtain a multi-channel foreground target image; determining the high-level features of the multi-channel foreground target image, the high-level features including: high-level features of the image of the battery cell sealing ring to be inspected, high-level features of the defect-free front image, and high-level features of the defect-free back image; and performing adversarial classification between the high-level features of the image of the battery cell sealing ring to be inspected and the high-level features of the defect-free front image and the defect-free back image, respectively, to obtain a probability value of the battery cell sealing ring image to be inspected containing surface defects.

The utility model discloses a kind of active infrared detection devices of wind power blade, the device includes inspection robot, six degrees of freedom mechanical arm, detection executor module;Detection executor module includes pedestal, laser heater, infrared thermal imager, laser range finder, laser heater is fixed in the middle of pedestal, irradiation direction is vertically downward, infrared thermal imager is located laser heater left side, infrared thermal imager is fixed in the bottom of pedestal by support inclination, laser range finder is located laser heater right side, laser range finder is vertically downward setting.The utility model carries out wind power blade defect detection using active infrared thermal imaging technology, heating is carried out to wind power blade by laser heater, the temperature distribution of blade surface is detected using infrared thermal imager, to find the defect possibly existing in blade interior.

This invention belongs to the field of communication automation technology and discloses a method for implementing a high-reliability HBM bonding process. The invention covers the entire HBM bonding process from wafer pretreatment and precise bonding to post-processing stress control by constructing three core modules: precise pretreatment of the bonding surface, high-precision bonding of heterogeneous wafers, and dynamic stress control after bonding. Employing core technologies such as precise detection and repair of micro-defects on the bonding surface, sub-micron-level alignment of heterogeneous wafers, and dynamic calculation and compensation of bonding interface stress, it overcomes the industry bottlenecks of traditional HBM bonding processes, including "difficulty in controlling bonding surface defects, low alignment accuracy, and easy interface stress failure." It achieves high-cleanliness pretreatment of the HBM bonding surface, sub-micron-level precise bonding of heterogeneous wafers, and dynamic adaptation and control of interface stress after bonding, significantly improving the yield, interface bonding strength, and long-term operational reliability of HBM bonding, thus meeting the advanced packaging requirements of next-generation high-bandwidth memory chips.

The embodiment of the specification discloses a kind of dynamic pseudo-color rendering methods, equipment and medium of large image data, it is related to image rendering technical field, for solving the problem of low reading efficiency and low rendering precision of current large image data.Method includes: obtaining the source image data generated by semiconductor optical detection;Based on the resolution level corresponding to different scaling ratio, pyramid segmentation is carried out to source image data, and mixed tile data and its corresponding level row column index are constructed;Mixed tile data includes: raster base map tile data and vector overlay tile data;According to the scaling level and display area coordinates of current view, the corresponding level row column index is determined to load the corresponding target tile data;Target tile data is input into pre-training mapping model, and the corresponding pseudo-color mapping value generated by the pre-training mapping model is obtained, to be rendered and displayed based on pseudo-color mapping value.

The utility model discloses a novel equipment for lens visual inspection, its technical scheme's main point is including frame, be equipped with carousel through five work position divider drive on the frame, be equipped with five work positions for placing lens on the carousel, be equipped with lens feeding and blanking device on the first work position, be equipped with first industrial camera and first highlight bowl shape light source for detecting lens lower part respectively on the second work position, be equipped with second industrial camera and second highlight bowl shape light source for detecting lens upper part respectively on the third work position, be equipped with third industrial camera, first line light source, first motor for detecting lens upper part respectively on the fourth work position, be equipped with fourth industrial camera, second line light source, second motor for detecting lens lower part respectively on the fifth work position, the industrial code reader for scanning the code to lens to mark whether to be the qualified product. Realized the full automation, high efficiency, high accuracy's multidimensional detection.

The invention provides an automatic testing method and device based on test sequence generation, equipment and a medium, and relates to the technical field of server testing. The method comprises the following steps: acquiring natural language test requirement data, and generating a structured target test step sequence based on a test requirement; then resource state information of the environment to be tested is collected, the resource state information is combined with the target test step sequence, and a test task scheduling instruction used for indicating a step execution strategy is generated; and distributing the test task scheduling instruction to a plurality of distributed execution nodes in a to-be-tested environment, and executing test steps in the scheduling instruction by the distributed execution nodes to obtain a corresponding test execution result. According to the scheme, automatic generation from natural language requirements to executable test steps can be achieved, reasonable scheduling is carried out based on the environment resource condition, and the execution efficiency and accuracy of the test task in the distributed environment are improved.

The utility model discloses a chip package auxiliary detection device, include: bearing base, its top fixedly connected with the mounting seat cylinder, the inside fixed mounting of mounting seat cylinder has the lift cylinder, the telescopic end top fixed mounting of lift cylinder has the bearing connecting plate, the upper surface fixed mounting of bearing connecting plate has visual detection camera no.

The embodiment of the application provides a kind of based on dynamic feedback's organic electroluminescence uniformity detection method and system, belong to organic electroluminescence uniformity detection field;The method comprises after initializing detection system, and calibrating reference baseline, acquisition standard plate reference image;Design first sparse sampling template;First sampling is carried out using first sparse sampling template and reconstruction algorithm, and the reconstruction full-width image of organic electroluminescence device under each waveband is obtained, and first light emitting uniformity evaluation is carried out;Mark abnormal light emitting area, dynamically adjust sampling strategy, using second sparse sampling template carries out second sampling, updates the reconstruction full-width image under each waveband, and carries out second light emitting uniformity evaluation, and exports organic electroluminescence uniformity detection report.The application is improved by intelligent sparse sampling and closed-loop feedback technology Organic electroluminescence uniformity detection efficiency and accurately locate defect area, provide efficient solution for production quality control.

The invention discloses an austenite welding seam ultrasonic detection method which comprises the following steps: detection preparation: selecting a bicrystal longitudinal wave angle probe matched with the acoustic performance of a detected austenite welding seam, a reference block and a coupling medium; instrument calibration: performing time-base line, probe angle and sensitivity calibration on the ultrasonic instrument by using the standard test block, and establishing a distance amplitude curve based on the lateral transverse hole of the reference test block; the base material in the detected area is detected through a normal probe, secondary back face echoes are adjusted to 75% of the full-screen height, and the defect that bottom waves disappear is recorded; and welding seam scanning: scanning a welding seam and a base material or a heat affected zone in a range of at least 10mm on two sides by using the bicrystal longitudinal wave angle probe, and covering the whole detection volume. The method has the beneficial effects that the technical scheme optimizes the characteristics of the austenite welding seam, and the defect detection rate is increased by more than 20% compared with that of a traditional method; and the positioning error is less than or equal to 1.5 mm.

The application provides a method and system for testing the inference ability of an artificial intelligence system. The method comprises: constructing a multi-dimensional space probability distribution model based on a structured configuration file of a plurality of dimensional parameter atomic items and at least one variation strategy rule defined for an inference ability scenario of an AI model deployed on an AI system; generating a dynamic test case set based on the multi-dimensional space probability distribution model; and executing a test on the AI model deployed on the AI system using the dynamic test case set.

The invention belongs to the technical field of LED display screens, and particularly relates to an LED display screen lamp bead missing automatic detection system based on machine vision, which comprises a conveying line, a detection part is arranged on the outer surface of a frame of the conveying line, the detection part comprises a detection camera, and the detection camera detects lamp bead missing of the LED display screen. According to the LED display screen lamp bead missing automatic detection system based on machine vision, flexible self-adaptive fixing and lossless clamping of an LED display screen are achieved through the arrangement of the driving component and the clamping component, specifically, the lifting frame is driven by the driving device to enable the contact air bag to make contact with the screen body, filtered clean gas is inflated, and then the LED display screen lamp bead missing automatic detection system is achieved. The multiple contact air bags can be attached to the surfaces of the display screens with different thicknesses or slightly deformed in a self-adaptive mode, pressure is evenly dispersed, and micro cracks caused by screen body pressing damage, scraping or stress concentration possibly caused by a traditional rigid clamp are effectively avoided.

A kind of automobile engine factory inspection test method, belong to engine power test technical field.The method will engine be installed to test bench, carry out appearance inspection, fluorescent agent addition, combustion chamber endoscopy and cylinder pressure detection in turn, then complete running-in test according to four-stage gradient working condition, monitor ECU operation data and listen to abnormal sound during running-in period, detect leakage using fluorescent flashlight after running-in, can be factory shipped after all detection is qualified.The present application effectively solves the problem that production line cannot detect by standard, complete bench test process, realizes comprehensive, efficient, stable engine factory inspection, and effectively improves engine factory shipment quality and detection reliability.

The invention discloses a test scene intelligent simulation method, system and device for a real-time control system and a medium, and the method comprises the steps: receiving a test demand, carrying out the analysis of the test demand based on a large language model and a test domain knowledge graph, and generating a structured test demand; calling a corresponding generative artificial intelligence model, and generating a cross-modal test data set associated in time sequence and semantics; based on a structured test requirement, a cross-modal test data set and a test domain knowledge graph rule, automatically generating a test script and arranging a time sequence test scene process so as to drive the real-time control system and collect response of the real-time control system; and analyzing a test execution result, optimizing decision logic of data generation or protocol simulation through reinforcement learning based on the analysis result, and updating the decision logic to the test domain knowledge graph. Compared with the prior art, full-link automation of real-time control system testing from scene construction and execution to autonomous evolution is achieved.

This invention discloses a non-destructive testing method based on a multidimensional complementary ensemble empirical mode decomposition (MEMD) algorithm. The method constructs a three-dimensional matrix of all thermal image data and transforms it into a two-dimensional matrix. MEMD is then performed on each row of the two-dimensional matrix, and the decomposed images are further processed column-wise. The resulting images are then combined using the MEMD algorithm to extract the defective image portion, yielding the clearest image. This invention utilizes both the temporal and spatial information of the thermal image sequence through row and column mode decomposition, comprehensively reflecting defect information. Furthermore, this invention has broad applicability, suitable for any thermal image sequence, including both long-pulse heating and eddy current pulse heating.

The invention discloses a steel rail welding seam flaw detection scanning device and a flaw detection method thereof. The steel rail welding seam flaw detection scanning device comprises a scanning support, a guide rail sliding block mechanism and a guide rail probe mechanism, wherein the scanning support can be mounted on a steel rail in a falling manner and is automatically positioned on the steel rail or is automatically separated from the steel rail through lifting; the guide rail sliding block mechanism comprises a first guide rail arranged in the length direction of the steel rail, a guide rail sliding block arranged on the first guide rail and a sliding block driving mechanism used for driving the guide rail sliding block to reciprocate along the first guide rail, and a rail surface probe capable of being attached to the rail surface of the steel rail is fixed to the bottom of the guide rail sliding block. The second guide rail is located below the first guide rail and is parallel to the first guide rail, the probe support is arranged on the second guide rail in a sliding mode, a rail side probe is fixed in the probe support, the guide rail sliding block can drive the probe support to reciprocate along the second guide rail, and the guide rail probe mechanism and the scanning support are of a linkage structure. The steel rail welding seam flaw detection scanning device and the flaw detection method based on the same are convenient and efficient to operate.

The utility model relates to the technical field of aluminum casting detection, in particular to an aluminum casting appearance defect detection device which comprises a workbench, a feeding mechanism, a conveying mechanism and a defect detection mechanism, the feeding mechanism, the conveying mechanism and the defect detection mechanism are arranged on the workbench, the feeding mechanism is used for automatically conveying aluminum castings, and the conveying mechanism is used for transferring the aluminum castings among stations; and the defect detection mechanism is used for detecting the aluminum casting on the carrier. The double-manipulator cooperative system is adopted, the detection efficiency is improved, the universality is enhanced, the cost is remarkably reduced, and the intelligence and the expandability are improved, so that the existing technical problems are effectively solved.

The invention provides an automatic regression testing method and device suitable for a risk control system, and relates to the technical field of real-time risk monitoring. The method comprises the steps of collecting transaction flow data of multiple channels of a test environment; customizing a test case including a risk control rule according to the actual risk control transaction scene; embedding a simple factor, an index factor and a third-party factor; and configuring a test case according to the scheduling dimension, performing timed automatic regression execution of the test case based on the simple factor, the index factor and the third-party factor to obtain a test result, and generating a test report according to the test result. According to the embodiment of the invention, data pre-burying of various complex scenes such as the simple factor, the index factor and the third-party factor is supported, the test case can be executed in a timed and automatic regression manner based on the simple factor, the index factor and the third-party factor, the test result is obtained, the test report is generated according to the test result, the defect detection rate is increased, and the delivery quality is improved; meanwhile, the human cost of regression test execution is reduced, and the regression test efficiency is improved.

This utility model relates to a clamping chuck for a wheel hub X-ray inspection device, belonging to the field of non-destructive testing technology for wheel hubs. The chuck includes a coaxially assembled top cover, a toothed disc, a conical base, and a drive shaft. The top cover, toothed disc, and conical base each have a coaxially aligned center hole and a set of circumferentially distributed positioning holes. A keyway and key fit achieve circumferential fixation between the drive shaft, toothed disc, and conical base. Positioning pins passing through the three sets of positioning holes achieve phase alignment and synchronous rotation of the top cover, toothed disc, and conical base. The upper surface of the toothed disc has conical teeth, the lower part of which forms an annular groove with the upper surface of the conical base to clamp the lower rim of the wheel hub. The conical teeth press against the wheel hub bead seat. The chuck body is made of 45# steel to ensure structural strength and wear resistance. The drive shaft is driven by an external drive assembly, driving the entire chuck and the wheel hub to rotate precisely.

This utility model discloses a device for detecting internal defects in hinge beams, comprising a dual-station alternating detection component, including a first-station detection platform and a second-station detection platform symmetrically arranged on the left and right, and an embedded centering and alignment mechanism; a detection component, including a six-axis robotic arm and internal detection probes located at the end of the six-axis robotic arm that scan the outer surface of the hinge beam one by one according to a preset path trajectory; and a three-axis travel mechanism, including a lateral travel component for switching between the two stations and for the lateral movement of the six-axis robotic arm, a longitudinal travel component for the longitudinal movement of the six-axis robotic arm, and a vertical travel component for the vertical movement of the six-axis robotic arm. This utility model adopts a dual-station alternating detection mode, realizing parallel operation of detection and loading / unloading. When the hinge beam at one station is being scanned and detected, the operator can simultaneously complete the unloading and clamping operations at the other station. The three-axis travel mechanism drives the detection component to move between the two stations, eliminating equipment waiting time.

The invention provides a system and method for detecting and classifying rail defects by using artificial intelligence, and the system comprises a receiving module which is used for inputting one or more rail images; the preprocessing module is used for preprocessing the rail image; the Transform module is used for carrying out defect detection and false alarm analysis on the preprocessed image by adopting a two-stage Transform model; and the output module is used for outputting and displaying a result containing defect, misinformation and unknown information. The method has the advantages that the method is easy to implement, the false alarm rate of an existing railway track detection system can be reduced, and the defect detection rate is increased. The method is easy to implement, and the risk of accidents or injuries in the defect detection process is reduced. And the service life of the track is prolonged, and the requirement for large-scale maintenance or replacement is reduced to the maximum extent.