Industrial Machine Vision Systems Research Articles

Research on the Method of Light Intensity Detection based on Kalman Filtering

A light intensity detection algorithm using Kalman filtering is proposed for the usage of data fusion in multi-sensor scenarios, like industrial control, machine vision, and navigation systems. In traditional measurements, the result is often faced with noise interference, causing accuracy problems. This algorithm Kalman filtering can be a good solution to noise signal reduction and provide optimal estimation in various domains that need prediction. As Kalman filtering is suitable for light intensity measurements, This work experimented with the steps shown below. First, the light intensity data is reported by sensors, and the mean and standard deviation of the measurement data are calculated; Second, by introducing a Kalman filtering model into the detecting model, the precision of estimating is improved, and the reliability of the system is enhanced; Finally, the data and diagram are exported by Matlab, and the Kalman filtering algorithm is compared with classical algorithms like moving average, median filtering. The experimental data shows Kalman filtering can be better applied to state estimation and signal processing domains, and the average light intensity data is 949.744. This result supports that Kalman filtering provides a robust and efficient method and it bed-covers the future research in improving the precision in light intensity measurements.

Evaluating the depth resolution of 3D sensors for manufacturing automation applications

The Intelligent Systems Division (ISD) at the National Institute of Standards and Technology (NIST) is performing research to support the development of documentary standards within ASTM subcommittee E57.23 on Industrial 3D Machine Vision Systems. This subcommittee is addressing the performance evaluation of 3D imaging systems that are used for manufacturing automation and vision guided robotics. The applications of these systems extend to other areas such as robotic bin picking, autonomous vehicles, 3D reconstruction, and entertainment. This paper discusses the activities of one work item (WK73176) that is addressing the evaluation of the depth resolution of a 3D imaging system. The key contributions of this paper include the description of multiple methods to determine the resolution of a 3D imaging system.

Ensemble Learning Approaches to Data Imbalance and Competing Objectives in Design of an Industrial Machine Vision System

Numerous industrial applications of machine learning feature critical issues that need to be addressed. This work proposes a framework to deal with these issues, such as competing objectives and class imbalance in designing a machine vision system for the in-line detection of surface defects on glass substrates of thin-film transistor liquid crystal displays (TFT-LCDs). The developed inspection system composes of (i) feature engineering: extraction of only the defect-relevant features from images using two-dimensional wavelet decomposition and (ii) training ensemble classifiers (proof of concept with a C5.0 ensemble, random forests (RF), and adaptive boosting (AdaBoost)). The focus is on cost sensitivity, increased generalization, and robustness to handle class imbalance and address multiple competing manufacturing objectives. Comprehensive performance evaluation was conducted in terms of accuracy, sensitivity, specificity, and the Matthews correlation coefficient (MCC) by calculating their 12,000 bootstrapped estimates. Results revealed significant differences (p < 0.05) between the three developed diagnostic algorithms. RFR (accuracy of 83.37%, sensitivity of 60.62%, specificity of 89.72%, and MCC of 0.51) outperformed both AdaBoost (accuracy of 81.14%, sensitivity of 69.23%, specificity of 84.48%, and MCC of 0.50) and the C5.0 ensemble (accuracy of 78.35%, sensitivity of 65.35%, specificity of 82.03%, and MCC of 0.44) in all the metrics except sensitivity. AdaBoost exhibited stronger performance in detecting defective TFT-LCD glass substrates. These promising results demonstrated that the proposed ensemble approach is a viable alternative to manual inspections when applied to an industrial case study with issues such as competing objectives and class imbalance.

Reducing Pseudo-error Rate of Industrial Machine Vision Systems with Machine Learning Methods

Nowadays machine learning and artificial neural networks are hot topic. These methods gains more and more ground in everyday life. In addition to everyday usage, an increasing emphasis placed on industrial use. In the field of research and development, materials science, robotics and thanks to the spread of Industry 4.0 and digitalization, more and more machine learning based systems introduced in production. This paper gives examples of possible ways of using machine learning algorithms in manufacturing, as well as reducing pseudo-error (false positive) rate of machine vision quality control systems. Even the simplest algorithms and models can be very effective on real-world problems. With the usage of convolutional neural networks, the pseudo-error rate of the examined system reducible.

CAOS-CMOS camera

Proposed and experimentally demonstrated is the CAOS-CMOS camera design that combines the coded access optical sensor (CAOS) imager platform with the CMOS multi-pixel optical sensor. The unique CAOS-CMOS camera engages the classic CMOS sensor light staring mode with the time-frequency-space agile pixel CAOS imager mode within one programmable optical unit to realize a high dynamic range imager for extreme light contrast conditions. The experimentally demonstrated CAOS-CMOS camera is built using a digital micromirror device, a silicon point-photo-detector with a variable gain amplifier, and a silicon CMOS sensor with a maximum rated 51.3 dB dynamic range. White light imaging of three different brightness simultaneously viewed targets, that is not possible by the CMOS sensor, is achieved by the CAOS-CMOS camera demonstrating an 82.06 dB dynamic range. Applications for the camera include industrial machine vision, welding, laser analysis, automotive, night vision, surveillance and multispectral military systems.

Design and application of industrial machine vision systems

In this paper, the role and importance of the machine vision systems in the industrial applications are described. First understanding of the vision in terms of a universal concept is explained. System design methodology is discussed and a generic machine vision model is reported. Such a machine includes systems and sub-systems, which of course depend on the type of applications and required tasks. In general, expected functions from a vision machine are the exploitation and imposition of the environmental constraint of a scene, the capturing of the images, analysis of those captured images, recognition of certain objects and features within each image, and the initiation of subsequent actions in order to accept or reject the corresponding objects. After a vision system performs all these stages, the task in hand is almost completed. Here, the sequence and proper functioning of each system and sub-systems in terms of high-quality images is explained. In operation, there is a scene with some constraint, first step for the machine is the image acquisition, pre-processing of image, segmentation, feature extraction, classification, inspection, and finally actuation, which is an interaction with the scene under study. At the end of this report, industrial image vision applications are explained in detail. Such applications include the area of automated visual inspection (AVI), process control, parts identification, and important role in the robotic guidance and control. Vision developments in manufacturing that can result in improvements in the reliability, in the product quality, and enabling technology for a new production process are presented. The key points in design and applications of a machine vision system are also presented. Such considerations can be generally classified into the six different categories such as the scene constraints, image acquisition, image pre-processing, image processing, machine vision justification, and finally the systematic considerations. Each aspect of such processes is described here and the proper condition for an optimal design is reported.

A real time marking inspection scheme for semiconductor industries

In this paper, a real time industrial machine vision system incorporating optical character recognition (OCR) is employed to inspect markings on integrated circuit (IC) chips. This inspection is carried out while the ICs are coming out from the manufacturing line. A TSSOP-DGG type of IC package from Texas Instruments is used in the investigation. The IC chip markings are laser printed. This inspection system tests whether the laser printed marking on IC chips is proper. The inspection has to identify print errors such as illegible characters, missing characters and upside down printing. The vision inspection of the printed markings on the IC chip is carried out in three phases, namely, image preprocessing, feature extraction and classification. The MATLAB platform and its toolboxes are used for designing the inspection processing technique. Speed of the marking inspection is mostly dependent on the effectiveness of the feature extraction technique. The performances of four feature extraction techniques are compared in terms of their respective speed. The feature extracted data are used in a neural network for classifying the marking errors. A suggestion to optimize the number of input neurons of the neural network for a fast classification is also presented.

LEDs: A flexible option for machine vision

Describes the benefits that arise from the use of Light Emitting Diodes (LEDs) for illumination in industrial machine vision systems. These include long life, stability and narrow frequency bands. Also discusses precautions that need to be taken for obtaining a uniform illumination and pulsed operation.

Time delay and integration imaging for inspection and profilometry of moving objects

Dynamic inspection of moving objects is a significant task in industrial machine vision systems. A CCD camera operating in time delay and integration (TDI) mode is equivalent to a digital drum camera, providing possibilities for dynamic inspection of moving objects (either rotating or translating on industrial platforms). We report some advances in the application of TDI cameras for the inspection and profiling of moving objects. The method is based on conventional light projection systems that are widely used in industry. The TDI system generates an on-line scanning moire´ display that could assist operators to identify defects on objects such as dents. Conventional projection systems use mechanical or interferometric types of fringe projections, which is not desirable in the industrial environment. The TDI system makes use of temporal modulation of a laser diode, which could be easily controlled by a PC or by the output ports in vision systems. In addition to the details of system implementation, we also describe two algorithms for unwrapping the deformation or shape profile of the test object based on the fast Fourier transform and sinusoidal fitting of the recorded patterns.

Prototype machine vision system for segmentation of hide images

AbstractWe give a brief description of the operation principles of LeaVis, a prototype industrial machine vision system aimed at processing and segmentation of the images of hides marked by lines and other symbols that show defects and areas of different quality. The goal of LeaVis is to provide a visual input for a CAD system that sets trajectories for the knife cutting the hides. We discuss the basic motivations of the vision system design, present some of its algorithms, and demonstrate examples of their operation.

Machine vision for industry: Tasks, tools and techniques

The ‘why’, ‘how’ and ‘what’ of industrial machine vision systems are surveyed-why vision is important, how it is accomplished and what sorts of tasks it is being applied to. Examples are given of vision techniques and applications from Japan, France, the GDR and the USA.