Machine Vision Inspection Research Articles

Defect Detection of Scroll Fixed Using AI Machine Vision Inspection

Defect Detection of Scroll Fixed Using AI Machine Vision Inspection

Research on the Detection Method of the Threshing Rate of Corn Ears Based on Machine Vision

The threshing rate is one of the important indexes to evaluate the effect of corn threshing. The weighing method is often used to calculate the depuration rate of maize at present. This method is time-consuming and laborious and can only calculate the overall threshing rate but does not give the threshing rate of individual corn ears. Different parameters of corn ears have complex effects on the threshing rate. By analyzing the threshing rate of each corn ear, we can choose the appropriate ear treatment method, optimize the processing equipment and process flow, and improve the threshing performance. This paper presents a method based on machine vision to detect the threshing rate of corn ears. In this method, machine vision was used to measure the parameters of the corncob and the area of the top of residual kernels. The area of the top of all kernels was restored based on the parameters of the corncob. The threshing rate of corn ears was calculated by the ratio of the area of the top of the missing kernel to the area of the top of all kernels after threshing. A bivariate linear regression area model was established to restore the area of the top of all corn kernels based on corncob parameters. The R2 was more significant than 0.98, and the goodness of fit was good. The machine vision inspection results showed that the maximum relative error of length and midsection radius was 7.46% and 5.55%, and the mean relative error was 2.58% and 2.23%. The maximum relative error of the corn ear threshing rate was 7.08%, and the mean relative error was 2.04%. When the residual kernels were concentrated in the midsection, the inspection result of the corn ear threshing rate was better. The maximum relative error was 3.98%, and the mean relative error was 1.07%. This paper provides a new idea and reference for measuring the threshing rate of corn ears.

Recent Patents for Machine Vision Inspection in the Machine Tool Sector

Abstract:: In the current modern manufacturing industry, the demand for product quality control is increasing, as is the need for automation. However, the manual inspection may need to be noticed or undetectable in the complex production process. To solve this phenomenon, machine vision-based inspection of machine tools and their product features has become one of the key technologies in intelligent manufacturing. The basic principle of machine tool machine vision inspection is to transmit the object to be inspected to the computer through efficient image acquisition. Then, image processing or deep learning algorithms classify, locate, segment, and process the features of the thing. Finally, the computer makes the judgment of whether it is qualified. According to recent numerical analysis data, compared with traditional methods, machine tool machine vision inspection can improve the inspection efficiency by 30%, and the accuracy has reached 98%, much higher than manual inspection accuracy. This study provides an in-depth review of representative patents and machine tool vision inspectionpatents , and it comprehensively summarizes the research results from various dimensions in recent year. In the past five years, the average annual growth rate of machine tool vision inspection technology has reached 15%, showing the rapid development momentum of this field. The research content mainly focuses on designing vision detection structures, algorithm optimization designs, and personalized solutions for object differences.

Deep convolutional neural network for weld defect classification in radiographic images

The quality of welds is critical to the safety of structures in construction, so early detection of irregularities is crucial. Advances in machine vision inspection technologies, such as deep learning models, have improved the detection of weld defects. This paper presents a new CNN model based on ResNet50 to classify four types of weld defects in radiographic images: crack, pore, non-penetration, and no defect. Stratified cross-validation, data augmentation, and regularization were used to improve generalization and avoid over-fitting. The model was tested on three datasets, RIAWELC, GDXray, and a private dataset of low image quality, obtaining an accuracy of 98.75 %, 90.255 %, and 75.83 %, respectively. The model proposed in this paper achieves high accuracies on different datasets and constitutes a valuable tool to improve the efficiency and effectiveness of quality control processes in the welding industry. Moreover, experimental tests show that the proposed approach performs well on even low-resolution images.

Machine Vision for Knot Detection and Location in Chinese Fir Lumber

Abstract In order to be utilized in the design of a wood building, the lumber must pass grade. Machine-vision inspection grading offers higher efficiency and accuracy than traditional manual visual grading. In this paper, a fast and accurate method for identifying defects in large-size structural lumber based on machine vision of Fujian Chinese fir (Cunninghamia lanceolata (Lamb.) Hook) constructional lumber (FCF CL) is proposed. Specifically, the grey matrix of the captured images on the surface of the sawn timber is initially scanned and the pixel weights on the edges of the image greyness variables are calculated. A matrix-valued torus was formed by fitting the knot edge profile and analyzing changes in the gradient values at the knot's edge, as well as calculating the directional derivative's rate of change. The knot three-dimensional mapping curves were projected onto the plane to form horizontal rise contours. Observe from the contour map of the whole large-size sawn timber, and extract the positional information of the knot where there is a trough (groove). The test results show that the rRMSE (Relative Root Mean Square Error) measured at the x axis position of knots is within 0.49 percent; the rRMSE measured at the y axis is 0.35 percent, which has high detection accuracy and meets the production requirements. We also investigated the effect of knots in different positions on the modulus of elasticity and the bending strength of FCF CL, with a view to establishing a link between machine-vision knot detection and mechanical properties of sawn timber in our next work.

Simulation and modeling of grinding surface topography based on fractional derivatives

The grinding surface’s 3-D topography has self-affinity. To more accurately simulate this characteristic, this paper proposes for the first time to construct a grinding surface simulation model based on fractional derivatives, including: utilizing the Weierstrass-Mandelbrot function to simulate the point cloud of the grinding wheel surface; further deriving the trajectory equation based on the He’s fractal derivative; the semi-inverse method is employed to construct the fractional-order variational principle of the vibration equation. Experiments show that the simulation model designed in this paper can predict the surface height more accurately, and is better than the traditional model in the calculation of roughness parameters Sa and Sq. The model is helpful to thoroughly understand the formation mechanism of grinding surface topography, and provides simulation data guidance for machine vision inspection of grinding surface in micron scale.

A novel deep learning algorithm applied to machine vision inspection for surface defects of injection moulded products

In modern industry, the surface defect inspection of injection moulded products is crucial for controlling product quality and optimising the manufacturing process. With the development of optical measurement and computer technology, machine vision inspection methods have been widely adopted instead of manual inspection. However, current machine vision inspection methods are difficult to simultaneously ensure the accuracy and efficiency of surface defect inspection of injection moulded products. Considering this problem, a novel deep learning algorithm applied to machine vision inspection for surface defects of injection moulded products is proposed. To train and evaluate the proposed deep learning algorithm, an image acquisition platform is established and the dataset of surface defects in moulded products is obtained. In the proposed deep learning algorithm, reparameterisation-based convolution modules are employed for feature extraction and feature fusion. A median iterative clustering algorithm based on hierarchical clustering initialisation is proposed to obtain prior anchors that are highly matched with the actual distribution of defect sizes. A novel focus-entire union over covering loss function is utilised for bounding box regression. On these bases, the proposed deep learning algorithm applied to machine vision inspection is evaluated on the dataset of surface defects in moulded products. The experimental results indicate that compared to the traditional inspection algorithms and other deep learning algorithms currently used in machine vision inspection, the proposed deep learning algorithm exhibits superior inspection accuracy and inspection efficiency on the acquired dataset. The inspection precision reaches 0.964, the inspection recall reaches 0.955, and the inference time for each subgraph is only 6.1 ms, confirming its effectiveness.

Quantitative analysis of the effect of illumination variations on image processing in machine vision inspection

Abstract When unavoidable light, age, expression, and gesture changes occur, the machine vision detection accuracy will be greatly reduced, especially the light change impact on image processing. Combining the reflection equation and frequency domain analysis to construct a light change model based on spherical harmonic function and process the image from four directions, namely, digital image conversion, median filtering, sharpening processing, and image global segmentation. Aiming at the shortcomings of traditional image processing algorithms, the light compensation based on the improved pulse-coupled neural network model is proposed, and the model is used to analyze the impact of light changes on image processing. Despite the changes in light conditions, this algorithm maintains above 0.95 in each subset and has the best overall performance, as shown by the results. When the light intensity is less than 41.88kLux, using the optimal threshold for background segmentation of adult green peppers with respect to a fixed threshold T=0.3 results in a significant reduction of the background segmentation error. When the light intensity is greater than 41.88kLux, the superiority of the optimal threshold is not obvious. The results of this research not only have a significant role in promoting the development of the field of computer vision but also can provide people with convenient and efficient services in all aspects of social life.

Automated Machine Vision Inspection System for Applying Plastic Lids on Wet Wipes Packs

Computer vision techniques have been widely used in industrial manufacturing for automation, monitoring, quality assessment and inspection. Specifically, some techniques are being deployed to detect and identify products’ defects instead of human operators. In this paper, a vision system is proposed to inspect and assess the quality of applying plastic lids on wet wipes packs at a manufacturing company in Jordan. Currently, the manufacturing company uses automatic lid applicator that has no visual control system and thus, a human operator is needed to inspect every pack. This process includes detecting the position and orientation of each lid to assure that it has been applied correctly at the center of the pack above specific label. Automating this inspection process would save time and efforts and increase the productivity. The proposed inspection system has the following modules: (1) Pack and lid detection and segmentation using YOLOv8s-Seg algorithm; (2) distance to border (DtB) extraction between the pack centroid and lid boundary; and (3) inspection module using linear Support Vector Machine (SVM). A segmentation dataset of 319 different images of wet wipes packs was constructed using the Segment Anything Model (SAM). The proposed inspection has been tested on different wet wipes packs. Experimental results successfully demonstrated the efficiency of the inspection system.

Research on defect detection of bottle cap interior based on low-angle and large divergence angle vision system.

During the machine vision inspection of the inner section of bottle caps within pharmaceutical packaging, the unique conca bottom and convex side walls often create obstructions to the illumination. Consequently, this results in challenges such as irregular background and diminished feature contrast in the image, ultimately leading to the misidentification of defects. As a solution, a vision system characterized by a Low-Angle and Large Divergence Angle (LALDA) is presented in this paper. Using the large divergence angle of LED, combined with low-angle illumination, a uniform image of the side wall region with bright-field characteristics and a uniform image of inner circle region at the bottom with dark-field characteristics are obtained, thus solving the problems of light being obscured and brightness overexposure of the background. Based on the imaging characteristics of LALDA, a multi-channel segmentation (MCS) algorithm is designed. The HSV color space has been transformed, and the image is automatically segmented into multiple sub-regions by mutual calculation of different channels. Further, image homogenization and enhancement are used to eliminate fluctuations in the background and to enhance the contrast of defects. In addition, a variety of defect extraction methods are designed based on the imaging characteristics of different sub-regions, which can avoid the problem of over-segmentation in detection. In this paper, the LALDA is applied to the defect detection inside the cap of capsule medicine bottle, the detection speed is better than 400 pcs/min and the detection accuracy is better than 95%, which can meet the actual production line capacity and detection requirements.

Optimization study of anomaly detection algorithm in machine vision inspection technology

Abstract In recent years, target detection algorithms based on machine vision have been a hotspot in computer vision research. The You Only Look Once (YOLO) algorithm, as an excellent target detection algorithm, has played an important role in improving detection speed and accuracy with the improvement of the network architecture in its development process. This paper introduces the concept of integrated learning to the YOLOv5 network architecture, incorporating deformable convolution and attention mechanisms. It also chooses the focal EIOU loss function to replace the GIOU loss function, thereby addressing the issue of localization loss, prioritizing abnormally detected targets, and enhancing the detection efficiency of these abnormal targets. Finally, we examine the practical value of the improved YOLOv5 algorithm by testing its performance and applying it to real-world anomaly detection. The results show that the improved YOLOv5 model outperforms the original YOLOv5 model in terms of performance and practical application advantages. In terms of performance, the classification accuracy of sea_person and earth_person in the improved YOLOv5 model is 37% and 26%, respectively, which is a significant performance gain overall. In actual application tests, the proposed method is more accurate than the traditional method. The accuracy is significantly higher.

Practical Applications of a Set-Based Camera Deployment Methodology.

This work establishes a complete methodology for solving continuous sets of camera deployment solutions for automated machine vision inspection systems in industrial manufacturing facilities. The methods presented herein generate constraints that realistically model cameras and their associated intrinsic parameters and use set-based solving methods to evaluate these constraints over a 3D mesh model of a real part. This results in a complete and certifiable set of all valid camera poses describing all possible inspection poses for a given camera/part pair, as well as how much of the part's surface is inspectable from any pose in the set. These methods are tested and validated experimentally using real cameras and precise 3D tracking equipment and are shown to accurately align with real imaging results according to the hardware they are modelling for a given inspection deployment. In addition, their ability to generate full inspection solution sets is demonstrated on several realistic geometries using realistic factory settings, and they are shown to generate tangible, deployable inspection solutions, which can be readily integrated into real factory settings.

Geometric Imperfection Simulations in Cee-Shape Cold-Formed Steel Members Based on Newly Developed Machine-Vision Inspection Techniques

This paper insightfully studies geometric imperfection simulations of cee-section CFS members from laser-based measurements. A machine-vision imperfection inspection technique is first developed where an algorithm is implemented to automate imperfection characterization from volumetric data. The measured imperfections are statistically analyzed and resemble past measurements from other researchers. Two imperfection simulation approaches are studied, i.e., the modal imperfection simulation method and the 1D spectral simulation method, where strength performance and deformation are predicted from finite element analysis. The analysis results are compared with those of testing. The 1D spectral simulation method is superior where stochasticity and regularity of real im-perfection can be properly addressed. The study provides feasible access to imperfection simulations of cee-section CFS members that other researchers can directly apply. The prediction results can aid the future direct analysis and design of CFS members.

Research on Surface Flaw Detection of Industrial Components Based on Deep Learning Algorithm

This paper primarily focuses on the research of industrial component surface defect detection algorithms using deep learning techniques. Taking the example of defect detection in the injector valve seat of automotive engine components, currently, the inspection of such small components heavily relies on manual visual inspection or traditional machine vision methods. Manual visual inspection is inefficient and cannot guarantee the speed and accuracy of detection. Traditional machine vision methods, while efficient, are sensitive to harsh environmental conditions and can be influenced by factors such as lighting, camera positioning, and background, resulting in poor robustness and limited feature extraction capabilities. To address these issues, this paper conducts research on a deep learning-based surface defect detection algorithm for injector valve seats.

Inspection of improved ghost imaging for hight-light LED with different factory quality

High brightness and high reflective properties are two important challenges for machine vision inspection. We propose a new method for nonlinear feature extraction and inspection of visual weak signals from highlight-emitting rather than “interference” bright objects. Based on the ghost image principle, the highlight characteristics of the light source become a useful characteristic. Under the guidance of ghost image theory, optimized design of ghost lens group, visual inspection of bright objects for the purpose of obtaining ghost images with high signal-to-noise ratio. We improved the high-brightness LED appearance quality inspection experiment of ghost image. Results show that the inspection method of improved ghost image of highlighted objects combined with deep learning has a good detection result. The method has an accuracy rate of 90.1%. In the early stages of the visual inspection process, there is great potential in applying the theory and method of ghost image to detect objects with high brightness and high reflective properties.

Cotton Gin Stand Machine-Vision Inspection and Removal System for Plastic Contamination: Auto-Calibration Design

Plastic contamination in marketable cotton bales, predominantly from plastic wraps used in John Deere round module harvesters, poses a significant challenge to the U.S. cotton industry. Despite rigorous manual efforts, the intrusion of plastic into the cotton gin’s processing system persists. We have developed a machine-vision detection and removal system aimed at mitigating this problem. This system employs inexpensive color cameras to detect plastic on the gin-stand feeder apron and subsequently removes it, reducing contamination. This system, built around low-cost ARM computers running Linux, comprises 30–50 machines and requires considerable effort to calibrate and tune. Moreover, its operation represents a technological challenge to typical cotton gin workers. This research presents a solution to this calibration operational hurdle by introducing an auto-calibration algorithm that has potential to simplify the system into a plug-and-play device. The auto-calibration system is designed to dynamically track the cotton color and utilizes frequency statistics to avoid plastic images that could compromise the system’s performance if incorporated into the auto-calibration process. We detail the design of the auto-calibration algorithm, which is expected to significantly reduce the setup overhead and facilitate the system’s continuous use. This innovation minimizes the need for skilled personnel and, therefore, is expected to expedite the system’s adoption across the cotton ginning industry.

Application and validation of machine vision inspection for efficient in-process monitoring of complex biomechanical device manufacturing

A technique is presented for shifting the manufacturing quality control of complex biomechanical catheters away from destructive testing of finished parts. This technique uses a more efficient real-time in-process monitoring through the application of machine vision inspection of patient critical quality parameters. The approach acknowledges the challenge of this industry operating in a strict regulated environment. The higher standards of built-in quality are achieved by developing automated inspection solutions that are more accurate and repeatable. Machine vision system and associated inspection job tools are developed and used to detect defects at crucial stages of manufacturing. The vision system is then tested for its robustness using a statistical approach to ensure its measurement capability is within the allowable process range and tolerances. The integrated solution developed is proven to be robust and highly precise in maintaining the manufacturing process stable. It enabled the manufacturing process to move away from a destructive double sampling plan with a standard LTPD of 5% to an otherwise real-time 100% non-destructive verification of units. This technique provides an alternative to otherwise cost-inefficient quality control inspections utilized in regulated manufacturing environment. It gives confidence to these conservative industries to move towards adopting digital manufacturing and Industry 4.0 practices.

Machine vision inspection of early failure and line-shaped defects of blue InGaN/GaN light emitting diodes soaked in liquid nitrogen for cryogenic tests

Cryogenic tests were conducted on blue InGaN/GaN light-emitting diodes (LEDs) by immersing the device in liquid nitrogen (LN2). To study the degradation of the LEDs, a combination of optical, electrical, and material characterizations were performed on the LN2-soaked device. The results indicate that noticeable emission spectral shifts were observed during the LED immersion due to lattice deformation. Moreover, clustering of cracks was generated in the cross-section, and line-shaped defects appeared on the surface of the GaN LED, which was different from the defects on AlInGaP LEDs after LN2 immersion. To track the trend of device damage, computer-assisted OpenCV-Python software was used to zoom in on nuanced variations in certain areas of the emission images. Early degradation of certain areas can be detected by computer-assisted machine vision, which has the potential to provide an early warning and protective circuits for the system consisting of the initial degraded LEDs. Reliability studies on LN2-soaked InGaN/GaN LEDs show promise in studying the reliability issues of GaN device applications in Arctic regions and space exploration.

Light Source Modules for Defect Detection on Highly Reflective Metallic Surfaces

High strength and highly reflective metal sheets are widely applied in industry; industrial requirements for defect detection are extremely demanding, particularly in the aviation and automotive industries. Classifying and recognizing surface defects on steel plate surfaces is highly challenging. Currently, defect detection is still inspected visually by personnel. However, given the high temperatures at inspection sites and the high risks in the operating environments, machine vision inspection systems are expected to replace manual inspection processes eventually. Therefore, this study developed an automated defect detection system that reduces the high reflectivity of examined objects. The light sources emitted light rays to the rays diffused and reflected multiple times inside the hemispherical cover to produce uniform illumination. Subsequently, image processing was conducted to highlight defect features on the stainless-steel plates. Relatively favorable light source positions were identified, which reduced the difficulty of class identification, the omission rate in defect detection to be decreased, and frequently encountered reflection problems in the automated optical inspection of metallic products to be overcome.

차량용 크래쉬패드 스티치 불량 검출을 위한 실시간 머신비전 검사 방법

In response to the market’s need for luxurious automobile interiors, automotive parts makers are developing various types of crash pads to give drivers a sense of emotional luxury. In particular, a low-cost and high-quality crash pad manufacturing technology is being developed for mid- to low-priced vehicles, namely, the IMG-S (In Mold Grain-pre Stitch) technology. High defect rate of stitching is a critical problem during the manufacture of crash pad using the IMG-S technology. In order to solve this problem, this paper proposes a method of real-time machine vision inspection of stitches on the automotive crash pad. This paper presents the real-time machine vision inspection system configuration, proposes stitch and reference line detection methods, and method for calculating the distance between stitches and the reference line. According to the distance between the stitch and the reference line, the status of the stitch was judged as normal, warning, or erroneous, and the final result was displayed on the user interface. The applicability of the proposed real-time machine vision inspection method was verified by stitching the test line.