39 results about "High speed vision" patented technology

The invention discloses microminiature needle body high-speed visual screening equipment and a screening method thereof, and belongs to the technical field of industrial automatic detection and sorting. The equipment comprises an equipment frame, a feeding module, a visual identification module, a grabbing module, a control module and a material receiving module. The feeding module is used for conveying to-be-screened needle bodies. The visual identification module performs image acquisition and model identification on the needle body; the grabbing module is provided with a plurality of grabbing clamping jaws, can move in the horizontal direction and the vertical direction under the driving of the control module, and grabs a plurality of needle bodies of the same model at the same time at a time. The material receiving module comprises a plurality of material receiving boxes and a transposition driving mechanism for driving the material receiving boxes to move so as to align the target material receiving box to the grabbing module. The equipment disclosed by the invention can be widely applied to high-efficiency sorting of micro-miniature needle bodies with the diameters ranging from 0.3 mm to 3mm and the lengths ranging from 3mm to 77mm, has high precision, high speed and high reliability, and is suitable for large-scale production and multi-variety and small-batch production requirements.

The invention relates to the technical field of crossing of artificial intelligence and intelligent manufacturing, discloses a position self-adaptive AI regulation and control method and system for continuous heat treatment of a fastener, and aims to solve the problems of non-uniform temperature field, high hardness dispersion, increase of quenching cracking rate and the like caused by lack of position perception and dynamic compensation lag in the prior art. According to the method, in-furnace coordinates of fasteners are tracked in real time through a high-speed visual positioning device, a thermal demand prediction neural network model with space-time convolution mixed with a gating circulation unit is constructed in combination with the conveying speed, thermal environment parameters and material characteristics, and an instantaneous thermal power instruction is output; the system comprises a high-speed visual positioning module, a motion state monitoring module, a thermal environment sensing module, a thermal demand prediction module and the like, and drives a high-density partitioned infrared heating array to realize millisecond-level local thermal compensation. According to the method, the temperature deviation of the junction of the temperature zones can be controlled within + / -8 DEG C, the hardness dispersion and the cracking rate are remarkably reduced, and the heat treatment consistency and the finished product yield are improved.

The invention discloses an additive manufacturing spatter behavior monitoring and drop point prediction method based on high-speed vision and fluid-solid coupling simulation, and belongs to the field of metal additive manufacturing monitoring. The method comprises the following steps: acquiring an image sequence of a selective laser melting process through a high-speed camera, carrying out image preprocessing, precisely segmenting splashing particles by adopting a combined morphological algorithm comprising Canny edge detection, expansion and corrosion operation, and acquiring the particle size, the initial velocity and the spraying angle of the splashing particles through multi-frame tracking; the initial velocity and the spraying angle are calculated by combining a displacement vector of the same particle in adjacent frames with a frame interval and a pixel calibration coefficient; afterwards, the parameters serve as initial conditions and are input into a particle dynamics model considering the influence of a protective airflow field and nonlinear air resistance based on a Schiller-Naumann model, a flight path is deduced through numerical integration, and a drop point of the flight path on a substrate is predicted; and finally, risk grade judgment is performed according to drop point distribution, and guidance is provided for process parameter optimization.

The invention discloses an unmanned aerial vehicle control method based on multi-sensor fusion, and relates to the technical field of unmanned aerial vehicle control, and the method comprises the steps: collecting a target image through a high-speed vision camera carried on an unmanned aerial vehicle; processing the target image based on an improved adaptive threshold algorithm, and determining a landing platform of the unmanned aerial vehicle; calculating horizontal deviation data, vertical height data, yaw angle data, pitch angle data and roll angle data of the unmanned aerial vehicle relative to the landing platform; acquiring the speed and acceleration of the unmanned aerial vehicle in real time; on the basis of the relative position prediction model, relative pose information of the unmanned aerial vehicle and the landing platform is updated in real time through an interpolation algorithm; and performing flight path planning according to the relative pose information, and controlling the unmanned aerial vehicle to land on a landing platform in stages. According to the method, the real-time performance of unmanned aerial vehicle control is improved, the control delay of the unmanned aerial vehicle is effectively reduced, and the control precision of landing of the unmanned aerial vehicle based on the landing platform is improved.

The present invention provides a high-speed visual scanning system and method based on a liquid lens, belonging to the field of visual scanning technology. The system includes: a relationship determination module, which is used to obtain multiple images of a scene to be scanned at the same time based on a liquid lens, and determine the mapping relationship between the actual shooting range and depth data of each image; a scene fusion module, which is used to perform scene fusion on all images based on the mapping relationship of each image, and obtain a fused image and a scene fusion strategy; a sequence generation module, which is used to generate a high-frame image sequence of the scene to be scanned based on the fused image and the scene fusion strategy; and a visual scanning module, which is used to determine the high-speed visual scanning results of the scene to be scanned in the high-frame image sequence based on deep learning; and is used to achieve high-speed and accurate scanning of objects contained in the scene to be scanned.

The invention discloses a thickness detection system and method, belongs to the technical field of transformer core lamination thickness detection, and particularly relates to a lamination thickness detection system and method for a transformer core production line, and the system comprises a beat monitoring module, a lamination thickness detection module, an upper computer, and a wireless communication module. According to the invention, the lamination thickness data of each step can be effectively detected in the automatic iron core lamination process. Through high-speed visual processing, a 3D laser sensor and a mobile robot technology, a detection position is rapidly and accurately positioned, the production takt is followed, and a production line is helped to find and process abnormal data in time, so that the production efficiency and the product quality are improved, and an important support is provided for automatic and intelligent development of a production process.

The application relates to an industrial-vision-based barite ore sorting and distributing device and a sorting machine, and belongs to the technical field of ore sorting. The device comprises a supporting frame and a conveying belt assembled in the supporting frame. A high-precision visual detector, a dynamic hot film mechanism and a high-speed visual sensor are sequentially arranged on one side of the supporting frame from bottom to top. The optimized design of the dynamic hot film mechanism realizes efficient water and dirt removal of wet and sticky ores. Two groups of ultrasonic transducers are arranged on the upper and lower sides of the hot air cavity, the vibration direction of the ultrasonic transducers is at a specific angle with the axis of the hot air flow, and the forced air flow formed by the built-in heating rod and the high-pressure vortex fan can make the ultrasonic vibration and the hot air flow act on the ore surface in cooperation, so that the surface moisture can be quickly evaporated through heating, and stubbornly adhered impurities can be peeled off with the help of ultrasonic vibration. Meanwhile, the dustproof net at the output end of the air outlet bin can prevent the backflow of impurities, guarantee the cleanliness of the ore surface after pretreatment, and provide a clear detection object for optical detection.

The utility model relates to visual inspection technical field, and disclose copper wire film surface quality high -speed vision detection system, including device frame, the right side of device frame is equipped with the detection assembly for to copper wire surface detection, the inside of device frame is equipped with the cleaning assembly for to copper wire cleaning, the detection assembly includes mounting piece, mounting piece is equipped with two, two mounting piece all weld in the right side of device frame, the utility model discloses through vision sensor collection image, after gauss filter denoising, histogram equalization enhances image details, the image after processing can accurately identify scratch, bubble, wrinkle, film thickness uneven etc. defect type based on the semantic segmentation model U of depth learning Net, MaskR CNN, and the detection precision reaches 0.01mm 2 , and further effectively detect the defect of copper wire, and further reduce the subsequent processing of the copper wire with defect.

The invention discloses silicon carbide non-pressure barrel surface finishing and grinding equipment and method, and belongs to the technical field of precision ceramic machining. The method comprises the following steps: cooperatively obtaining a high-precision three-dimensional point cloud on the surface of a cylinder; macroscopic geometric deviation identification based on three-dimensional registration and microdefect intelligent identification based on a PointNet + + deep learning model are executed in parallel, and a structured defect feature set is generated; according to the defect features, a layered contour grinding instruction for macroscopic protrusions and a fixed-point spiral grinding instruction for microscopic pore cracks are generated through self-adaptive planning; and when the machining instruction is executed, the posture of the barrel body is monitored in real time through a high-speed vision sensing system, the movement track of the grinding head is dynamically adjusted in a closed-loop mode, and system errors are compensated. By constructing an intelligent closed loop of perception-decision-execution-compensation, integrated accurate identification and self-adaptive high-precision grinding of macro and micro composite defects on the surface of the silicon carbide cylinder are achieved, the technical problems that a traditional method is not accurate in identification, the strategy lacks adaptability, and errors cannot be compensated in real time are effectively solved, and the method is suitable for large-scale popularization and application. And the machining precision and the automation level are obviously improved.

This invention proposes a spatiotemporally synchronized visualization measurement and analysis method, system, and device for flow-heat-phase. The method includes: acquiring a multimodal sensing dataset and performing spatial calibration and alignment; processing the aligned high-speed visual image to obtain a binary phase diagram, combining temporal information to identify and segment it, dividing and merging the liquid phase region, gas phase region, contact line region, and rewetting region to obtain a partitioned result map; identifying and fusing the infrared temperature image, binary phase diagram, and partitioned result map to output a spatiotemporally synchronized visualization measurement phase diagram for flow-heat-phase. By acquiring data to form a dataset and calibrating and aligning it, this invention achieves, for the first time, a precise spatiotemporally synchronized correlation of gas-liquid distribution, temperature field, and heat flow information; combining binary phase diagram generation, intelligent segmentation, and multimodal data fusion technology, it overcomes the limitations of traditional methods and can quantitatively analyze key physical processes under the interaction of multiple bubbles on complex surfaces; the final output phase diagram can reveal the coupling mechanism between the temperature field of the micro-region of the heated surface and the dynamic behavior of bubbles.

This invention discloses a saw blade detection method and system based on multi-source data fusion, specifically relating to the field of saw blade detection technology. By simultaneously acquiring high-speed visual images of the saw blade's front and side views and spindle current signals, the initial tooth profile reference features of the saw blade are extracted. Furthermore, the system monitors in real time changes in the straightness of the projection line, the amount of cutting edge retraction, and abnormal changes in the current signal during the saw blade cutting process. By fusing visual and current data analysis, the normal wear and degradation trend of the saw blade and abnormal tooth loss characteristics are observed, thereby achieving accurate identification of saw blade wear or tooth loss conditions. A comprehensive evaluation is then performed based on the defect rate of sawn products, ultimately outputting an accurate and reliable saw blade replacement control signal. This invention effectively improves the accuracy of saw blade condition detection and the timeliness of replacement decisions, avoiding processing quality problems and ensuring production efficiency and processing safety.

The present invention discloses a production device for nanofiber yarn, comprising a fixed line end and a moving line end for fixing the two ends of the yarn, comprising a control mechanism for controlling the moving line end to move away from the fixed line end from slow to fast, the moving line end and the fixed line end can keep the yarn horizontal during detection, comprising a high-speed visual system located below the detected yarn, which is configured to record data at the yarn breakage point, preferably comprising an electric slide and an electric slider sliding on the top thereof, the top of the electric slider is fixed with a limit seat, the outer wall of the limit seat is slidingly connected to an intermediate block through a slide groove, the moving line end is installed on the intermediate block, the control mechanism comprises a servo motor installed on the top of the electric slider, the output end of the servo motor is fixedly mounted with a crank, and the crank is slidingly connected to a transmission rod by opening a slide groove. The present invention can improve the accuracy of detecting the amount of yarn stretching.

The present application relates to a kind of PCB drill bit micro-chipping online detection and early warning method.For the problem of difficult tool state identification under different materials and working conditions, a kind of high-speed vision, current, vibration and displacement signal is synchronously collected, a high-resolution multi-modal data set covering typical micro-chipping evolution stage is constructed, and based on normalized feature fusion, lightweight time coding and fingerprint database establishment, high-fidelity representation and online dynamic matching of micro-chipping state are realized.The system accurately identifies and predicts the evolution trend through dynamic time warping and Bayesian inference, and realizes adaptive early warning feedback loop by linking machining control system, while supporting incremental optimization of fingerprint database based on artificial review.The scheme effectively improves the sensitivity and discrimination accuracy of drill bit health state identification, and provides technical support for efficient and intelligent machining process monitoring.

The utility model discloses a four head high -speed visual labeling machine, including lower workbench, be equipped with label peeling group on lower workbench, product transport group, still be equipped with three -axis mechanical arm of along XYZ direction transport on the platform, wherein, three -axis mechanical arm includes the connecting plate, four groups of drive motor are equipped on the upper side of connecting plate back, the driving gear shaft that is connected with corresponding drive motor is equipped on the upper side of connecting plate front, the driven gear shaft is equipped below connecting plate front, driving gear shaft axle center and driven gear shaft axle center are perpendicular line, is equipped with vertical slide rail on this perpendicular line, and driving gear shaft and driven gear shaft are equipped with transmission belt, one side of transmission belt is equipped with with vertical slide that is assembled on vertical slide rail, the vacuum suction nozzle that can take air and release air is equipped on the vertical slide front, and the vacuum pump is externally connected to vacuum suction nozzle through the pipeline, and this technology is used to the small specific product, and adopts four head vacuum suction nozzle mode, satisfies specific product label and work efficiency.

The invention discloses a robot grabbing system and method based on computer vision, and relates to the technical field of industrial vision intelligence and robot control. The method comprises the following steps: acquiring a high-speed visual image sequence and multi-source disturbance data of the tail end part of a flexible target object in real time; performing spatial-temporal feature extraction on the image sequence to obtain end displacement spatial-temporal features; constructing a nonlinear dynamic model based on the multi-source disturbance data and the displacement spatial-temporal characteristics; taking the nonlinear dynamic model as a prediction model, and calculating an optimal control compensation sequence by solving a constrained optimization problem in a finite time domain; a control compensation generation instruction at the current moment is extracted, and the robot is driven to adjust the grabbing track; and repeating the steps to realize rolling optimization control. According to the method, the dynamic behavior of the flexible body is accurately described through the nonlinear dynamic model, active compensation of multi-source disturbance is realized in combination with model predictive control, and the technical problem of high-precision and high-stability grabbing of the flexible target in a complex environment is effectively solved.

The invention belongs to the technical field of artificial intelligence and robot control, and particularly relates to a handwriting trajectory self-adaptive correction method for a calligraphy practicing robot. The method aims at solving the problems that a traditional calligraphy practicing robot is difficult to adapt to individual writing styles, and trajectory correction intelligence is insufficient. The pen wielding track, the pressure, the speed and the pen tip inclination angle of a user are collected in real time through a high-density pressure sensing array and a high-speed visual sensor, and a multi-modal data stream is formed; a personalized handwriting feature model is constructed by adopting a long-short-term memory network, and dynamic features such as starting strength and turning acceleration are extracted; generating an initial expected track matched with a user style in combination with the standard Chinese character path; in the execution process, the inertia measurement unit and the electromagnetic positioning module are used for feeding back an actual track in real time. According to the method, the track naturalness and the correction accuracy are improved, the writing texture consistency is guaranteed, and multi-user concurrent and long-term personalized teaching is supported.

The invention discloses a self-adaptive lighting device for high-speed visual insect motion capture, which adopts an axial laminated integrated structure and comprises a control driving assembly, a multispectral light source assembly and a mixed optical assembly, the control driving assembly is electrically connected with the multispectral light source assembly, the multispectral light source assembly comprises an annular substrate and a plurality of light emitting units arranged on the annular substrate, and the light emitting units are distributed on the annular substrate in a multi-circle concentric ring shape. The red light chips, the green light chips and the blue light chips in the light emitting units are arranged in a staggered mode, the mixed optical assembly is located on a light emitting path of the multispectral light source assembly, and the mixed optical assembly comprises a double-face compound eye light uniformizing device. The contradiction between the high-frame-rate imaging requirement and the biological low-interference requirement is solved, and high-quality uniform illumination is achieved.

The invention discloses a real-time high-speed visual inspection system based on deep learning, and relates to the technical field of crossing of computer vision and industrial automation. Comprising multiple paths of GigE cameras, and the frame rate is greater than or equal to 500fps; the industrial controller carrying a GPU co-processing card realizes high-speed data exchange through a PCIe bus; a real-time task scheduling module based on Linux RT, wherein the priority is greater than 99%; a reasoning tool of a TensorRT engine is integrated, and dynamic input size adjustment is supported; and the image preprocessing module optimized by the CUDA realizes zero-copy video memory mapping. Through collaborative optimization of software and hardware, an NI PXIe-8881 controller and an NVIDIA T4 GPU co-processing card are adopted, and Linux RT real-time task scheduling and self-developed AI Vision Toolkit are combined, so that high-speed synchronous acquisition and low-delay reasoning (the total delay is 3.5 ms) of four cameras at 526.5 fps are realized, and the detection flux reaches 13000 + piece / min. The system supports dynamic input size adjustment, a wide-temperature environment and electromagnetic interference resistance, the deployment period is shortened to 3 days, and the contradiction among real-time performance, precision and stability in an industrial scene is solved.

The invention discloses a high-speed visual positioning method for highly reflective multi-material printed matters. According to the method, image acquisition is synchronously triggered through hardware, an adaptive algorithm based on local gray statistics is utilized to enhance an image, a lightweight double-branch neural network is adopted to realize sub-pixel-level mark positioning, and a motion model is combined to carry out time sequence filtering and then output. According to the method, an FPGA is adopted as a core processor, a complete algorithm assembly line is hardened in a chip, and a deterministic delay processing path from image input to control output is formed. The method effectively solves the industrial problems of instable identification of tiny marks in a complex optical environment and uncertain response delay of a software scheme, and is especially suitable for a high-speed and high-precision printing and packaging equipment visual positioning scene.

The invention relates to the field of suture welding, in particular to a laser-arc welding suture welding system based on visual feedback. According to the system, an initialized welding parameter set is obtained, based on the initialized welding parameter set, the welding process with the space-time coupling relation formed by laser beam pre-melting and pulsed arc filling in sequence is started, multi-mode sensing of high-speed vision and spectral signals is synchronously triggered, and a welding real-time process data set is generated; on the basis, fusion deconstruction analysis is conducted on the geometric shape of a molten pool, temperature field distribution, the relative position of laser and electric arc acting points and plasma physical characteristics, and a welding state characteristic set is generated; on the basis, the energy ratio, the space relative position and the action time sequence of the laser and the electric arc are dynamically regulated and controlled, and a self-adaptive control instruction set is generated and executed. In the laser-arc welding suturing welding process, the welding technology is promoted to be upgraded to automation and intelligence, and powerful support is provided for large-scale popularization of the laser-arc welding suturing welding technology.

The invention relates to the technical field of metal material processing, in particular to a metal material processing system which comprises a main processing unit, a multi-dimensional sensing unit, an autonomous moving material processing unit and a central control unit, and the central control unit comprises a signal fusion analysis module, a digital twinning module, a process dynamic optimization module and a collaborative scheduling module. The multi-dimensional sensing unit comprises a vibration sensor, an acoustic emission sensor and a high-speed vision module. According to the invention, the rejection rate is reduced, the consistency of the quality of batch machined parts is ensured, seamless cooperation of machining and logistics is realized, the equipment waiting time is reduced to the greatest extent, the overall equipment utilization rate is improved, the data processing pressure of the central control unit is effectively reduced, and the response speed of the system to real-time events can be improved; and the real-time performance and the stability of the system are enhanced.

The application discloses a kind of intelligent swing laser welding systems and methods of stator and rotor core, it is related to laser welding field, including swing laser welding module, multi-modal online detection module, intelligent control and feedback module and motion execution module, by optical coherence tomography, high-speed vision, multispectral sensing and infrared thermal imaging real-time acquisition melt depth, molten pool shape, splash, spectrum and temperature and so on Multidimensional characteristic signal.Based on the real-time identification of defect by CNN-LSTM hybrid architecture deep learning model, and through millisecond level feedback control dynamic adjustment laser power, swing parameters and protective gas flow, realize the closed-loop quality control of " perception-decision-execution ".Solved the welding defect frequent, thermal deformation difficult to control and quality detection lag and other problems in the prior art, significantly improve the welding yield and consistency of stator and rotor core, meet the mass production demand.

The invention discloses a tablet appearance defect online high-speed visual inspection system which comprises a protection box and further comprises a feeding mechanism, a conveying assembly, a turnover device, a detection assembly, a removing device and a collecting device, the conveying assembly is located in the protection box and arranged in the middle of the protection box, the feeding mechanism is arranged at one end of the conveying assembly, and the turnover device is arranged at the other end of the conveying assembly. The turnover device is arranged at the other end of the conveying assembly, the collecting device is arranged below the turnover device, the detecting assembly is arranged between the collecting device and the turnover device, the removing device is arranged between the collecting device and the detecting assembly, and the removing device is arranged on one side of the detecting assembly. Detection is efficient and accurate, the medicine quality can be guaranteed, and the compliance risk and the labor cost of an enterprise are reduced.

The invention relates to a TIG (Tungsten Inert Gas) automatic welding pulse phase control method based on time sequence molten pool recognition. A dynamic image of the molten pool is obtained through a high-speed visual sensor, and multi-dimensional feature quantization of the front edge curvature, the rear edge liquid-solid interface gradient, the center gray scale standard deviation and the like of the molten pool is achieved in combination with self-adaptive filtering and feature extraction. A lightweight CNN-LSTM hybrid network is utilized to output a normalized base metal melting index, a molten pool expansion rate and a fillable confidence coefficient in real time, and then a continuous molten pool dynamic phase angle is generated based on an arc tangent function model. After the phase angle and the current phase offset are mapped, FPGA hardware double buffering processing is carried out, microsecond pulse current phase adjustment is achieved, and the arc stability is improved through an S-shaped acceleration limiting measure. According to the scheme, the molten pool state sensing precision and the welding process control real-time performance are effectively improved, and collaborative optimization of welding stability and filling adaptability is achieved.

The utility model discloses book high -speed visual inspection equipment, include: area array detection section, including first detection line, be equipped with the area array camera for gathering book top surface, first side and second side image on first detection line top and both sides, line sweep detection section, including the interval area formed by second detection line and third detection line, be equipped with the line sweep camera for gathering book bottom surface, third side and fourth side image below interval area and both sides, the utility model discloses the combination of area array detection section and line sweep detection section carries out the detection to the six -sided appearance of book, can identify common appearance defect, also can add up the verification of specified two -sided appearance feature image such as high -grade defect of internal defect folding.

The invention provides a target detection method based on a high-speed vision camera, and belongs to the technical field of computer vision, and the method comprises the steps: collecting a target image through the high-speed vision camera carried on an unmanned plane; graying and de-noising the target image to obtain a pre-processed target image; dynamically adjusting a threshold based on an improved adaptive threshold algorithm, eliminating shadow interference in the preprocessed target image, and obtaining an optimized binary target image; contour detection and polygon approximation are carried out according to the optimized binary target image, target feature information is extracted, the matching degree of the target feature information and a preset visual identifier is detected, and the landing platform of the unmanned aerial vehicle is locked. According to the method, the target image can be effectively identified and processed in an uneven illumination, shadow shielding or complex background environment, so that the requirement of real-time detection in a dynamic environment is met.

The utility model discloses high-speed visual dimension detection equipment, which relates to the technical field of dimension appearance detection of lithium battery aluminum shells, and comprises a rack and belt conveying equipment arranged at the lower end of the side part of the rack, a pipe frame is fixedly arranged at the upper end in the rack, a bottom plate is fixedly arranged at the bottom of the pipe frame, and a discharge hole is formed in the bottom plate; a transplanting module is fixedly arranged at the upper end in the pipe frame, a lifting air cylinder is fixedly arranged at the bottom of a moving part of the transplanting module, a vacuum suction cup assembly is fixedly arranged at the tail end of an output shaft of the lifting air cylinder, a plurality of positioning mechanisms are arranged at the end, away from the discharging hole, of the bottom plate, and two symmetrically-arranged connecting plates are arranged at the upper end in the pipe frame; the bottoms of the two connecting plates are each provided with two symmetrically-arranged high-speed detection cameras through a fine adjustment mechanism, and an automatic code scanner is fixedly arranged on the side wall of the bottom plate. According to the utility model, the product can be quickly scanned and grabbed and is accurately detected under the high-speed detection camera, so that the detection efficiency and the detection precision are improved.

This invention discloses an adaptive layered collaborative optimization method for metal additive 3D printing, relating to the field of 3D printing. The method includes: simultaneously acquiring real-time morphology, temperature field distribution, and dynamic fluctuation parameters of the molten pool during the metal additive 3D printing process using a high-speed vision acquisition module and an infrared temperature measurement module; performing noise reduction preprocessing on the acquired multi-dimensional raw data of the molten pool, simultaneously extracting core features, and performing standardized quantization on the core features to generate a molten pool state vector. This invention can accurately perceive the molten pool state, quantify core features, adaptively generate layer thickness and matching process parameters, select the optimal solution through adaptation judgment, and dynamically adjust parameters throughout the process, thereby improving deposition stability and forming consistency, reducing printing defects, adapting to various metal materials, improving part precision and quality, reducing scrap rate, and optimizing printing efficiency.

The invention discloses high-speed CCD visual inspection equipment for small plastic products. The high-speed CCD visual inspection equipment for the small plastic products comprises a box body, an upper computer, a feeding mechanism, a rotary conveying mechanism, a deviation rectifying mechanism, a visual inspection mechanism and a discharging mechanism, the box body comprises a mounting part and an isolation part, the mounting part is arranged at the bottom of the box body, the isolation part is arranged at the top of the box body, and the isolation part covers the top side surface of the mounting part; a feeding station, a deviation rectifying station, a first detecting station, a second detecting station and a discharging station are sequentially arranged in the top side area of the rotating driving end of the rotating conveying mechanism in the rotating conveying direction. The visual detection mechanism comprises a first visual detection unit and a second visual detection unit. According to the high-speed CCD visual inspection equipment for the small plastic products, the first inspection station and the second inspection station are arranged, so that the decomposition of inspection tasks is realized, the limitation of the visual angle of a single camera is avoided, and the omission ratio is greatly reduced.