Visual Surface Inspection Research Articles

Model-Based Segmentation of Surfaces Using Illumination Series

<para xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> Automated visual inspection of surfaces plays an important role in the context of industrial production. Segmentation is a key method in image processing of such surfaces. The appearance of structured surfaces depends very much on their illumination. Hence, we apply an illumination that is variable in its direction and in its shape. Image series are taken by varying the direction of the illumination pattern. The segmentation is performed on this data basis. We present an approach that utilizes the Torrance and Sparrow model to estimate the local reflection properties of the surface. The parameters of this model are then used as features to classify each surface point individually. </para>

Methodology for identifying particle sources in indoor environments

Growing concern about airborne particles in indoor environments requires fast source identification in order to apply remedial actions. A methodology for identifying sources emitting particles larger than 0.5 microm was designed and applied. It includes: (1) visual inspection of interior surfaces in order to identify deposited particles and inspection of potential sources (equipment, materials, activities etc.) of airborne particles. (2) Technical measurements of airborne particles at different positions in a building with simultaneous logging of activities. (3) Isolating potential activities/particle sources in a test chamber, initially free from particles, for controlled characterizations of the particles generated. The methodology was applied in a study of three houses in southern Sweden. The results show that source identification is facilitated by knowledge of concentration variations between different rooms, real-time measurements together with activity reports and information on particle characteristics that are comparable with results from laboratory simulations. In the houses to which the methodology was applied, major particle emissions from textile handling were identified, which were likely due to detergent zeolite residues.

Detection Functions for Archaeological Survey

This paper presents the results of several experiments to investigate how the detection functions of surveyors vary for different artifact types on surfaces with differing visibility when visual surface inspection (“fieldwalking”) is the survey method. As prospecting theory predicts, successful detection declines exponentially with distance away from transects and detection as a function of search time displays diminishing returns. However, these functions vary by visibility, artifact type, and other factors. The incidence of false targets–incorrect identifications of artifacts–has somewhat more impact at greater range but has little or no relationship with search time. Our results provide a rationale for selection of transect intervals and distribution of survey effort, and also facilitate evaluation of survey results, allowing more realistic estimates of how much a survey missed.

Virtual assembly of pottery fragments using moiré surface profile measurements

Reconstruction of archaeological monuments and potteries from fragments found at archaeological sites is a tedious task that requires many hours of work from archaeologists and restoration personnel. Excavations from famous archaeological sites in Jordan such as Petra, Jerash, Um Qeis, Ajlun etc. indicated civilization throughout centuries, from prehistoric periods, up to present. The restoration process is tedious and lengthy due to the mass of fragile fragments which consists of thousands of sherds found at excavation sites. An approach is the use of shadow moiré experimental technique to obtain the 3D model of pots from 3D measurements of the surface profile using the photos captured. This method is a non-destructive optical technique used for visual surface inspection, and surface profile measurements. The approach is simple, requires minimal mathematical calculations and reduces the risk of frequent handling of the fragile sherds, because the photos can be captured using a simple experimental setup in the site. Matching of fragments and aligning them geometrically is based on matching the profiles and edges of the broken parts virtually. This virtual assembly technique may assist the archaeologists as much as possible in solving their task with minimum time and minimum damages to fragments. The technique, of course, does not completely replace the archaeologist, but provides a useful estimation of valid fragment combinations, and accurately measures fragment matches. It also simplifies modeling the pots virtually using 3D-MAX modeling software to obtain the missing pieces.

Data requirements for identifying macroscopic water stress parameters: A study on grapevines

We have tested the inverse modeling approach to derive macroscopic water stress parameters (MWSP) using different types of information, such as soil water content, pressure head, and transpiration rate. This testing was performed by numerical experiments considering a multilayered soil growing grapevines under three different irrigation regimes and two contrasted water stress scenarios. The results indicate that measurements of the soil water content alone do not contain enough information to estimate MWSP. Nonuniqueness is likely to occur, and the MWSP estimates may contain large uncertainties. However, the incorporation of only transpiration measurements into the objective function does allow the deriving of accurate MWSP. This contrast is mainly due to the difference of sensitivity to the MWSP, which is much higher for the transpiration than for the soil water content. Results obtained using only soil water pressure head measurements are similar to or poorer than those obtained using transpiration data. Moreover, visual inspection of response surfaces of the objective function suggests that the incorporation of further information in addition to transpiration into the objective function is not of great value for the identification of MWSP. Uncertainties for the MWSP estimated using the three types of information combined are in most cases only 1.3 times smaller than when transpiration measurements alone are incorporated into the objective function. Beyond the specific results obtained for the estimation of MWSP we find that the parameters estimates and their associated uncertainties are strongly dependent upon the type, quantity, and quality of the information included into the objective function. Hence inverse modeling may provide a means to design better experiments.

Image acquisition techniques for automatic visual inspection of metallic surfaces

This paper provides an overview of three different image acquisition approaches for automatic visual inspection of metallic surfaces. The first method is concerned with gray-level intensity imaging, whereby the most commonly employed lighting techniques are surveyed. Subsequently, two range imaging techniques are introduced which may succeed in contrast to intensity imaging if the reflection property across the intact surface changes. However, range imaging for surface inspection is restricted to surface defects with three-dimensional characteristics, e.g. cavities. One range imaging approach is based on light sectioning in conjunction with fast imaging sensors. The second introduced range imaging technique is photometric stereo.

Automated metal surface inspection through machine vision

This paper proposes a method for automated visual inspection of metal surfaces. Firstly, the modified grey-level co-occurrence matrices of metal images are used to access the information of metal surfaces. Secondly, the difference moment and the entropy of the grey-level co-occurrence matrices are extracted as the features of the metal surfaces. Finally, the features of the inspecting images are then compared with the preset confidence interval to determine whether the inspecting metal is defective or not. Some combinations of relative positions between two positioning pixels and feature descriptors were tested in the experiments to find the best one. The experimental results show that the proposed method can detect the defects effectively and has better correct detection rates than the conventional method.

Real-Time Automatic Visual Inspection of High-Speed Plane Products by Means of Parallelism

Parallel processing can effectively satisfy the real-time constraints required by on-line machine vision applications. This paper describes a real-time automatic visual inspection (AVI) system for high-speed plane products, which is based on a reconfigurable and scalable coarse-grain distributed memory MIMD architecture, and a unique application programming interface. The code of application algorithms are source level machine independent. As our prototype is organized as a test bed, new algorithms and even new dedicated hardware processing elements can quickly be evaluated on it. The system can easily be tailored for various automatic visual surface inspection applications. An example for inspecting whether separate small mosaics (textures) in a scene are in normal shapes is described. The algorithm combines the connected component labeling, the moment calculation and the pattern recognition. This key task for most applications is well suited for the “divide and conquer” parallel paradigm. The real-time performance has been achieved on a TMS320C40 array.

Three dimensional automated visual inspection of surface mounted devices

Surface Mounted Technology (SMT) has created a demand for machine vision systems that can reliably verify component placement for screening inspection and process improvement. This paper describes a new 3-D computer vision system being developed at the Center for Electronics Manufacturing (CEM) of The University of Texas at El Paso (UTEP) for the inspection of Printed Circuit Boards (PCBs). A small range finder, utilizing a laser-based triangulation measurement method, is being employed for the development of an effective high resolution 3-D SMT inspection system.

Method for evaluation of imaging in automated visual web inspection

The quality of the image is crucial for the performance of a visual surface inspection system. The final image quality is affected by the properties of the imaging unit. In order to evaluate the quality of the imaging unit a designer should be able to characterize the imaging performance using quantitative figures of merit. The suitability of the modulation transfer function (MTF) for the figure of merit of imaging in visual web inspection is discussed. A straightforward setup for the determination of the system MTF is suggested. Three exemplary cases are analyzed: (1) the performance of an imaging system as a function of the system f number, (2) the performance of an imaging system against the depth-of-field requirement, and (3) the performance of a TDI imager against speed mismatching. In order to relate the system MTFs to image quality, a set of defective samples from a paper manufacturing process was collected and imaged. Quantitative image quality criteria are proposed, and the sample images are analyzed according to these criteria. The final image quality is compared with the corresponding system MTF values.

Automated Visual Inspection of Non-Smooth Surfaces. Application to Cast Aluminum

In this paper a general, low cost, arquitecture is described along with a methodologie for on-line defect detection and analysis in flat surfaces continously produced. On-line inspection of 100% production, along with the proper characteristics of inspection when texture is present force up the development of optimized algorithms for both detection and classification. Also hardware platforms that permit pipelined processing along with high modularity and are needed with a cost that make them competitive in every kind of surface visual inspection applications. An industrial application is described for quality assessment and process driving in cast aluminum obtention.

151 Automated visual inspection of textured surfaces

151 Automated visual inspection of textured surfaces

Fuzzy-logic tools on tap for IC wafers

The authors briefly describe the concept of fuzzy logic, and develop a technique for detecting dimple defects for automatic visual inspection of polished wafer surfaces. A fuzzy logic controller balances a beam in real-time without an explicit closed-form mathematical formulation. This in-house prototype can be operated automatically through the execution of a fuzzy control algorithm. Unlike most of the conventional and optimal control algorithms, this controller requires no system dynamics, and it is characterized by its simplicity and easy implementation. It is a convenient tool for automating operator controlled processes, although theoretical results concerning the system's stability and robustness are still not available and cannot be easily obtained. >

Automated visual inspection of rolled metal surfaces

A prototype for an automated visual on-line metal strip inspection system is described. The system is capable of both detecting and classifying surface defects in copper alloy strips, and it has been installed for evaluation in a production line in a rolling mill. The image acquisition part of the system is based on a CCD line scan camera and condensing bright field illuminators. The inspection algorithms are based on morphological preprocessing and combined statistical and structural defect recognition. The image processing hardware consists of commercial modules. An analysis of this implementation is presented. A similar inspection principle has also been successfully applied to steel strip inspection.

A Simple Interferometric Method for Workshop Testing of Optics

A simple step-by-step method is given for deriving the shapes of wavefronts from data obtained with a wavefront shearing interferometer. No mathematics, other than arithmetic, is used. The result is the accurate deviation of the wavefront from a reference sphere that coincides with it at three chosen reference points. The method is intended primarily for the use of opticians in optical workshops, but is also quite practical for the final testing of optics for performance rating. A method is given by which an optician can evaluate an optical surface by comparing the interferogram produced by it and a known prism interferometer, with a drawing of the desired interferogram. This procedure is analogous to using test plates for visual inspection of optical surfaces.