541 results about "Machine vision system" patented technology

Machine vision devices may be configured to automatically connect to a remote management server (e.g., a “cloud”-based management server), and may offload and / or communicate images and analyses to the remote management server via wired or wireless communications. The machine vision devices may further communicate with the management server, user computing devices, and / or human machine interface devices, e.g., to provide remote access to the machine vision device, provide real-time information from the machine vision device, receive configurations / updates, provide interactive graphical user interfaces, and / or the like.

The invention provides a textile defect detection method based on machine vision, and the method comprises the steps: obtaining a corresponding camera resolution parameter, an algorithm complexity parameter and a detection sensitivity parameter from a database according to a textile quality requirement, and carrying out the initialization setting of a machine vision system; performing feature extraction on the optimized textile surface image, extracting texture and morphological features related to wrinkle flaws, and constructing a feature spectrum capable of reflecting wrinkle features; identifying a suspected wrinkle flaw area in the characteristic spectrum, determining the contour of the flaw, and obtaining the position and range information of the wrinkle flaw; according to a pre-established wrinkle defect type database, performing mode matching on the quantized wrinkle characteristics, and if the matching degree exceeds a preset threshold value, determining the specific type of the wrinkle defect; and a support vector machine algorithm is adopted to classify the wrinkle degree quantitative indexes, wrinkle depth and density parameters are combined, different levels of wrinkle flaws are divided, and the wrinkle severity is evaluated.

The invention discloses an aluminum alloy surface defect detection method and system using deep learning, and relates to the related field of detection performed in combination with deep learning, and the method comprises the following steps: carrying out image acquisition on an aluminum alloy finished product, constructing a surface image data set to extract multi-scale surface features, and constructing a feature enhanced image set to carry out defect detection, and a defect detection result is obtained to dynamically optimize aluminum alloy production process parameters, and a process adjustment instruction is generated and fed back to a production system. The technical problems that traditional detection mostly depends on manual visual inspection or a machine vision system based on rules, imaging distortion, small defect missing detection and process adjustment are caused are solved, and the technical effects that deep learning is conducted through the double-branch network, the defect detection efficiency and the production yield are improved, and control over the aluminum alloy manufacturing quality is met are achieved.

The invention discloses an AOI optical scheme automatic optimization method based on reinforcement learning, and belongs to the technical field of automatic optical detection. According to the method, a reinforcement learning technology is applied to automatic adjustment of optical schemes in a machine vision system, wafer images under different optical schemes are collected by using an AOI system, and scoring is performed by using a plurality of semantic segmentation models; constructing and training a non-reference image quality evaluation network, evaluating the optical imaging quality in real time, and designing a reward function on the basis; by using the Actor-Critic algorithm, the intelligent agent can autonomously learn an optimal parameter adjustment strategy, quickly adapt to different scenes, realize dynamic optimization of light source parameters and remarkably improve image quality, and an efficient and intelligent solution is provided for efficient application of a machine vision system in a complex industrial environment.

The invention relates to the technical field of spraying process control, and discloses a spraying process self-adaptive adjustment method based on temperature measurement. The method comprises the steps that target coating parameters and base material physical property data are obtained, and initial spraying control parameters are generated through a first intelligent calculation model in combination with historical process data and real-time environment monitoring data; when spraying is executed, heat distribution data of a spraying area are collected in real time through a temperature sensing device, and according to the difference between the heat distribution data and a preset target temperature interval, technological parameters are dynamically corrected through a first self-adaptive regulation and control algorithm; meanwhile, a machine vision system is used for capturing actual coating morphological characteristics, and after the actual coating morphological characteristics are compared with target parameters, excitation parameters are adjusted in a partitioned mode through a second self-adaptive regulation and control algorithm so as to optimize uniformity; and collecting whole-process data, comprehensively evaluating the whole-process data through the second intelligent calculation model to generate a process optimization instruction, and updating the first intelligent calculation model and the first self-adaptive regulation and control algorithm parameters according to the process optimization instruction.

A machine vision system may include a meta-imager including a meta-optic and a polarization-sensitive photodetector. A machine vision system may further include and at least one processor operably coupled to the polarization-sensitive photodetector, and at least one memory operably coupled to the at least one processor. A machine vision system may be configured to: receive, from the polarization-sensitive photodetector, a plurality of feature maps; input, into a trained artificial neural network, the plurality of feature maps; and process, using the trained artificial neural network, the plurality of feature maps to recognize an object.

A measurement and positioning system based on machine vision and laser triangulation, which includes a machine vision system, a laser ranging system, and a standard positioning target configured to be installed on a positioned workpiece, where the laser ranging system is configured to project laser light onto the standard positioning target so as to perform a three-dimensional height measurement, and the machine vision system is configured to capture and image a positioning target and a laser image projected on the standard positioning target and to measure a two-dimensional coordinate position. The measurement system of the present application can realize a rapid three-dimensional measurement and precise positioning of a spatial position of the positioned object by means of the standard positioning target installed on the positioned object.

A system including: a robotic frame including at least an actuated arm, a machine-vision system including at least a camera, the machine-vision system communicatively connected to a computing device, the computing device including at least a processor and a memory communicatively connected to the at least a processor, the memory storing instructions configuring the at least a processor to: receive image data from the at least a camera, the image data including at least two captured viewpoints of a produce item, extract at least an object-level visual feature from the image data using a machine vision model, assign a defect classification label to the produce item as a function of the at least an object-level visual feature, and control the at least an actuated arm to perform a handling operation on the produce item as a function of the defect classification label.

The invention relates to the technical field of information, in particular to a forming process and system of a high-precision injection hook in the technical field of injection hook forming. The forming process includes the steps that computer aided design and finite element analysis are applied, the forming process of the injection hook in a mold is simulated, the mold structure is optimized, and the mold testing frequency is reduced; a high-precision mold is integrated with a cooling channel of a preset ejection hook arrow-shaped structure, a temperature sensor is arranged in the cooling channel, the cooling state is monitored in real time, cooling parameters are adjusted based on monitoring data, and precise structure forming is achieved. And the machine vision system detects the integrity and direction consistency of the arrow-shaped structure of the molded ejection hook on line, identifies unqualified products, automatically removes the unqualified products, records defect types, and sends defect type data and corresponding product batch information to the production data acquisition module.

The invention discloses a two-dimensional and three-dimensional fused fabric defect detection method, and belongs to the technical field of machine vision recognition. The technical problem of limitation of a traditional two-dimensional machine vision system is solved. According to the technical scheme, the method is characterized in that multi-angle light source imaging is carried out, a detected fabric is polished through light sources at different angles, and two-dimensional images of illumination at different angles are obtained through shooting of a linear array camera; and two-dimensional and three-dimensional information fusion: carrying out pixel-level fusion on the shot two-dimensional grey-scale map and the reconstructed gradient map and depth map with the three-dimensional information, carrying out high and low frequency decomposition and fusion on the two-dimensional information and the three-dimensional information through non-subsampled contourlet transformation, and fusing the two-dimensional defect feature information and the three-dimensional defect feature information into one map. After the non-subsampled contourlet transform decomposes the image, the frequency domain space is divided into wedge-shaped directional sub-bands; defects are identified based on the deep learning model. The method has the effect of improving the accuracy of textile quality detection.

A kaleidoscopic laser beam projector (KLBP) system that employs a single moving structure to project patterns of sweeping laser beams with diverse sweep directions and emission apertures. The system comprises a collimated light source, a spinning mirror deflector (“spindle / flutter mirror”) at the center, and a stationary kaleidoscopic mirror containing multiple facet mirrors arranged in a concave fashion around the spindle mirror. As the spindle mirror rotates, it directs laser beams in a radially whirling movement toward the facet mirrors, which then subdivide these beams into multiple short strokes and redirect them outward in different directions. Various embodiments include configurations with diffractive optical elements (DOEs) for beam splitting, different kaleidoscopic mirror geometries (hexagonal, octagonal, etc.), and innovative flutter mirror mechanisms that create controlled angular variations during rotation. The system enables a “4×1” configuration that replaces traditional “4×4” systems requiring eight motors and multiple moving structures. The KLBP system offers significant advantages including mechanical simplicity, reduced size and weight, improved energy efficiency, enhanced reliability, cost-effectiveness, and modular configurability. The diverse beam incidence angles enable superior illumination of complex three-dimensional objects for applications in machine vision systems, precision metrology, gap and flushness inspection, and robot-assisted assembly of complex structures.

The invention provides a weld quality evaluation system and method based on cooperation of ultrasonic waves and machine vision. The system comprises an ultrasonic detection system used for preliminary quality division and defect judgment of a weld and a machine vision system used for quality detection and quality evaluation of the weld. The ultrasonic detection system comprises an ultrasonic probe, an ultrasonic detector and a signal processing and analyzing module; the machine vision system comprises an image acquisition module, a data processing system and a user interface. The invention further provides a weld defect detection and evaluation method of the system. According to the method, ultrasonic detection and machine vision technologies are combined, ultrasonic waves are used for preliminarily detecting the defects of the welding seam, the defect detection process is gradually optimized, the machine vision system is combined for deep learning recognition of the surface defects of the welding seam, and finally independent operation of the machine vision system is achieved. Therefore, steel structure weld defect detection and weld quality evaluation are realized.

The invention discloses a building solid waste road performance index and multi-path utilization scene intelligent evaluation method. The method comprises the steps that the mass content of red bricks, the mass content of needle-sheet-shaped particles and the particle gradation in building solid waste particles are obtained through a machine vision system; the pavement performance indexes, namely CBR value, water absorption rate and crushing value, of the building solid waste are obtained through tests; according to the influence rules of the mass content of the red bricks, the mass content of the needle-sheet-shaped particles and the grain composition in the building solid waste particles on the building solid waste road performance indexes, an estimation model of the building solid waste road performance indexes is established. According to the method, the mass content of the red bricks, the mass content of the needle-sheet-shaped particles and the grain composition in the building solid waste particles are obtained in real time on the basis of machine vision, the road performance indexes of the building solid waste are predicted through the prediction model of the road performance indexes of the building solid waste, and the detection efficiency and accuracy are improved.

The invention relates to the technical field of surface defect detection, and discloses an aluminum profile surface defect analysis method based on machine vision, and the method comprises the steps: obtaining equipment operation parameters, environment parameters, aluminum bar processing procedure parameters and aluminum profile surface snapshots of processing equipment of a target aluminum profile in real time; obtaining a corresponding equipment operation risk value according to the equipment operation parameter; according to the aluminum bar machining procedure parameters, the equipment operation risk values and the environment parameters, procedure score values of the corresponding aluminum bar machining key procedures are obtained; the aluminum profile surface snapshots, obtained from multiple angles, of the target aluminum profile are combined and then sent to the pre-trained surface recognition model for evaluation, the quality inspection defect value output by the surface recognition model is obtained, and the surface quality defects can be continuously, accurately and rapidly found in a non-contact detection mode. The aluminum profile production efficiency and quality are fully improved, and compared with manual detection, a machine vision system has lower cost and higher efficiency.

Systems and methods for automatically adjusting the exposure settings (e.g., exposure time, analog gain, etc.) of imaging device(s) and / or camera(s) utilized in machine vision systems. According to embodiments of the present disclosure, one or more images or frames of a stream of images captured by an imaging device of a machine vision system can be processed and analyzed to determine exposure settings that may be applied to capture subsequent images by the imaging device. For example, a frame of the stream of images can be partitioned into a plurality of zones, and the pixels of each zone can be non-uniformly sorted into a histogram to represent the distribution of pixels in each zone. The histogram values for each zone can be aggregated to determine an aggregated exposure value associated with the frame, which can be compared against a target exposure value in view of a response function associated with the imaging device to determine the exposure value for capturing subsequent frames in the stream of images.

The invention relates to the technical field of mold processing, and provides a mold precision injection molding processing method and device based on machine vision, and the method comprises the following steps: carrying out all-directional scanning on the surface of a mold cavity based on a machine vision system to obtain cavity three-dimensional image data of the mold cavity; inputting the three-dimensional image data of the cavity into an image analysis and detection model to obtain a first analysis and detection result; calling a spectrum analyzer to detect the to-be-injection-molded material based on the first analysis detection result to obtain a material characteristic detection result of the to-be-injection-molded material; determining material processing parameters of the to-be-injection-molded material according to the material characteristic detection result, and melting the to-be-injection-molded material based on the material processing parameters to obtain a material melt; and first mold injection molding machining parameters are determined based on the mold characteristic parameters of the mold cavity, and the material melt is injected into the mold cavity based on the first mold injection molding machining parameters. The product quality stability and the production efficiency are improved, and the production cost is reduced.

The invention discloses a cigarette end surface defect detection method based on visual target detection, and the main design concept of the invention is that the method mainly comprises the steps: collecting original data of a cigarette end surface, carrying out the preprocessing of an original image, and segmenting the original image to obtain a cigarette end surface image; in combination with the scene characteristics of actually shooting the cigarette end face, expanding and reinforcing the cigarette end face image to obtain diversified cigarette end face visual samples; marking the cigarette end face visual sample and then training a target detection model based on deep learning; the trained target detection model is used for detecting a cigarette end face real-shot picture collected in the production process in real time, and the detected end face defects are subjected to quantitative characterization and then are subjected to score estimation. Compared with a traditional image processing method and a machine vision system, a new solution is provided for cigarette end face defect detection, and the dual requirements of a high-speed production line for detection precision and real-time performance can be met.

The invention discloses a double-arm citrus picking robot and a picking method, and belongs to the technical field of agricultural automation equipment. The robot comprises a crawler-type moving chassis, a symmetrically-arranged double-mechanical-arm system, a machine vision system, a citrus collecting bin and a control system. The double-mechanical-arm system adopts a main and auxiliary collaborative framework, the tail end of a main mechanical arm is integrated with a clamping jaw and vacuum suction disc parallel mechanism, an auxiliary mechanical arm is provided with a rotary circular saw blade, and lossless clamping of fruits and precise cutting of fruit stems are achieved. The machine vision system realizes high-precision citrus identification under complex illumination by improving a YOLOv5 network and fusing an HSV dynamic threshold value and Focal Loss optimization; graded path planning is combined with an A * algorithm and an RRT * algorithm, and motion safety of the mechanical arm is ensured through bounding box collision detection. The picking method comprises the steps of three-dimensional semantic map construction, collaborative grabbing and cutting, lossless transfer and the like. The problems that a single-arm robot is low in efficiency, fruits are prone to being damaged, fruit stems are not thoroughly treated and the like are solved, and the robot is suitable for automatic orchard harvesting in hilly areas.

The present invention relates to an integrated vision acquisition module for tunnel apparent defects, mainly used for solving the problems in existing detection devices of low integration level, complex debugging, low acquisition accuracy, poor environmental adaptability and the like. The vision acquisition module consists of a mounting back plate, a housing, a machine vision system, an integrated control system, a heat dissipation mechanism, a protective mechanism, and a wire arrangement mechanism. The machine vision system is encapsulated inside the mounting back plate and the housing, the heat dissipation mechanism is mounted on the back of the mounting back plate, the protective mechanism functions among the mechanisms of the vision acquisition module, and the integrated control system and the wire arrangement mechanism are mounted in the center of the front of the mounting back plate. The vision acquisition module uses a modular design, can implement accurate digital imaging of tunnel full-sections, has an overall protection grade of IP65 and good heat dissipation performance while maintaining a high degree of modularity, is adapted to different tunnel profile sections, and can complete rapid and non-contact detection of apparent defects in tunnel lining such as cracks, water leakage, and exfoliation and defects in ancillary facilities such as signal mechanisms, optical cables, and communication cables.

The invention relates to a high-precision material edge correction method. The method comprises the following steps: S1, deploying a machine vision system; s2, deploying a sensor network on a material transmission path; s3, preprocessing the collected material edge image, performing edge detection on the preprocessed image, and extracting an accurate contour of the material edge; an edge detection result and sensor data are accurately aligned and fused in time and space, and a multi-dimensional data set is obtained; s4, according to material characteristics and transmission system dynamics, establishing a physical model of material movement for predicting a future position of a material edge; s5, taking the multi-dimensional data set as the input of a physical model to train the physical model; s6, predicting the edge position of the material by using the trained physical model, comparing the predicted edge position of the material with the edge position of the material detected in real time, and calculating a deviation value; and generating a specific control instruction based on the deviation value, and correcting the deviation value.

The invention relates to the field of machine vision, in particular to a forklift obstacle avoidance method and system based on machine vision. Surrounding obstacles are recognized and positioned in real time through a machine vision technology, and the driving direction of the forklift is controlled, so that the driving path is dynamically adjusted, and the obstacles are effectively avoided. According to the invention, rich visual information can be provided through a machine vision technology, and complex and small object target obstacles can be identified more accurately. Compared with a single physical sensor, the machine vision system can ensure that the forklift can better adapt to changes of complex environments, and robustness and flexibility of the system are improved.

The invention relates to the technical field of endoscope optical performance testing, in particular to an improved endoscope optical performance testing method based on machine vision. The method converts the optical performance test of the endoscope from the traditional manual operation into automatic measurement, and specifically comprises the following steps of: measuring images of the endoscope under different test conditions by using a machine vision system so as to calculate various optical indexes of the endoscope; a mode of combining a Canny edge detection algorithm with a corresponding physical target is adopted, so that the calculation precision is improved; in the calculation process of indexes such as a field angle and an a value, a high-precision camera is used for analyzing imaging of a target, the calculation rate is improved by calculating geometric parameters and image features, and when sensitive indexes such as effective luminosity, unit relative distortion and a color rendering index are calculated, software is used for processing data obtained by a machine vision system, so that the calculation efficiency is improved. And personal errors are avoided.

A machine vision system for larval aquatic animal rearing constituted of a video camera, a watertight housing and a processor, wherein the video camera is arranged to continuously capture video of a predetermined volume and transmit the captured video to the processor, and wherein the processor is arranged to apply one or more neural networks to the captured video to: isolate individual aquatic animal within the video: identify at least one predetermined activity parameter and / or at least one predetermined morphological anomaly of the isolated aquatic animal.

A machine vision inspection system includes a movement portion comprising one or more movement mechanisms configured to adjust a relative position between an optical assembly portion and a workpiece. The optical assembly portion includes a camera that receives imaging light transmitted along an imaging optical path and provides images of a surface of the workpiece, and an objective lens that inputs the imaging light arising from the surface of the workpiece and transmits the imaging light along the imaging optical path to the camera. The optical assembly portion further includes a lens motion mechanism comprising a motion portion configured to enable motion of the objective lens from a rest position when a corresponding force is applied by a contact with the workpiece or other object (e.g., as corresponding to a collision with the workpiece or other object, and for which the system may stop motion when such a contact occurs).

The invention relates to a quick locking method and structure for an automobile part mold. The device is characterized by comprising a fixed base, a locking cylinder and a locking unit, in the motion control of the locking mechanism, a motion smoothing algorithm and a trajectory optimization algorithm are introduced, the motion trajectory of the locking mechanism is smoothed, the speed change and the acceleration change in the motion process are reduced, the vibration and impact of the locking mechanism are reduced, and the stability of the locking process is improved, so that the consistency of the locking position of the automobile part mold is ensured; and after the motion control is optimized, transmitting the optimized motion track data to a machine vision system for monitoring the subsequent locking process.

The invention discloses an appearance, flavor and taste coupled edible mushroom grading and sorting device and method. The device comprises a hardware part and a software part. Hardware comprises an airflow setting device, a conveying belt, an industrial camera, a light source, a pneumatic electromagnetic valve, an industrial personal computer and other main components. And software functions: while controlling the conveyor belt to rotate, acquiring images of the dried edible mushrooms by using the industrial camera, processing the images, obtaining the appearance quality of the edible mushrooms in real time, obtaining the content of main flavor components and the rehydration ratio of the edible mushrooms according to the appearance quality by using a built-in database, and outputting the grade of the dried edible mushrooms. And the corresponding electromagnetic valves of the corresponding grades are controlled to act, and the dried edible mushrooms are blown into the frames of the corresponding grades. According to the device, the machine vision system and the air-blowing type executing mechanism are used for grading the dried edible mushrooms, the appearance, the smell and the taste are combined, a multi-index fusion grading method is formed, and the grading accuracy and consistency are improved.

The invention relates to a flexible material cutting robot with a leveling function and a cutting method thereof, and belongs to the technical field of flexible material cutting, the flexible material cutting robot comprises a master control layer, a sensing layer, an execution layer and an interaction layer, the master control layer adopts a central control module, the sensing layer adopts a data acquisition unit, the execution layer is a hardware action unit, and the interaction layer adopts a man-machine interface. The central control module is provided with a digital twin engine and a multi-physics field collaborative optimization algorithm library, the data acquisition unit comprises a machine vision system and a material characteristic self-identification unit, and the hardware action unit comprises a multi-stage tension collaborative leveling module, a vacuum adsorption type dynamic cutting platform and a multi-axis linkage cutting head group. According to the invention, through closed-loop control of material characteristic sensing, dynamic leveling, self-adaptive adsorption and energy matching cutting, zero-deformation precise cutting of high-end flexible materials is realized, and the system is particularly suitable for high-value material processing in the fields of aerospace, medical equipment and the like.

Systems and methods are disclosed herein for identifying cable damage using machine vision, such as on a cable being fed down a wellbore from a cable spool in an oil and gas operation. An example method can include providing a camera directed toward a cable that is winding upon, or unwinding from. a cable spool. The method can also include capturing a plurality of frames of images of the cable by the camera. such as by capturing video. The captured frames can be cropped to remove portions of the frames that do not include the cable. The method can further include processing the cropped frames using a machine-learning model. The machine-learning model can be trained using images of known cable damage as inputs so it can identify new instances of damage. The machine-learning model can further classify each cropped frame as including damage or not including damage.

The invention provides a transplanting device with a sorting function. The transplanting device comprises a controller, a supporting mechanism, a seedling taking manipulator for grabbing pot seedlings, a moving mechanism for moving the seedling taking manipulator to a target position, a machine vision system for shooting pictures of the pot seedlings grabbed by the seedling taking manipulator, and a sorting manipulator for backwards moving the seedling taking manipulator for grabbing inferior seedlings, the collecting mechanism is used for collecting the remaining seedling taking manipulators after sorting, and the resetting mechanism is used for resetting the seedling taking manipulators. Each seedling taking manipulator is provided with an angle adjusting assembly and a distance adjusting assembly, the seedling taking manipulators can be clamped on the sliding rails or separated from the sliding rails by rotating the angle adjusting assemblies, and the distance between every two adjacent seedling taking manipulators can be adjusted through the magnetic attraction effect of the distance adjusting assemblies in the two adjacent seedling taking manipulators. By means of the transplanting device, the seedling taking mechanical arms grabbing poor-quality seedlings or empty seedlings can be moved backwards, only the seedling taking mechanical arms grabbing healthy seedlings are reserved for planting operation, the planting operation effect can be improved, and the transplanting quality is improved.

The present invention relates to the training of a convolutional neural network by a set of automatically generated simulated training data for use in a machine vision system for identifying fiducial tags associated with entities. Four-character fiducial tags are randomly generated or selected and then randomly transformed and superimposed on images to generate a set of simulated training data to be fed into a convolutional neural network to train the network model. The trained convolutional neural network is used to identify fiducial tags disposed on entities in real-world images and video.