Automatic inspection of metallic surface defects using genetic algorithms

This paper is concerned with the problem of automatic inspection of metallic surface using machine vision. An experimental system has been developed to take images of external metallic surfaces and an intelligent approach based on morphology and genetic algorithms is proposed to detect structural defects on bumpy metallic surfaces. The approach employs genetic algorithms to automatically learn morphology processing parameters such as structuring elements and defect segmentation threshold. This paper describes the detailed procedures which include encoding scheme, genetic operation and evaluation function.

The proposed method has been implemented and tested on a number of metallic surfaces. The results suggest that the method can provide an accurate identification to the defects and can be developed into a viable commercial visual inspection system.

Machine vision system for automatic inspection of surface defects in aluminum die casting

A camera based machine vision system for the automatic inspection of surface defects in aluminum die casting is presented. The system uses a hybrid image processing algorithm based on mathematic morphology to detect defects with different sizes and shapes. The defect inspection algorithm consists of two parts. One is a parameter learning algorithm, in which a genetic algorithm is used to extract optimal structuring element parameters, and segmentation and noise removal thresholds. The second part is a defect detection algorithm, in which the parameters obtained by a genetic algorithm are used for morphological operations. The machine vision system has been applied in an industrial setting to detect two types of casting defects: parts mix-up and any defects on the surface of castings. The system performs with a 99% or higher accuracy for both part mix-up and defect detection and is currently used in industry as part of normal production.

Machine vision system for automatic inspection of surface defects in aluminium die casting

A machine vision system is presented for the automatic inspection of surface defects in aluminium die casting. The system uses a hybrid image processing algorithm based on mathematic morphology to detect defects with different sizes and shapes. The defect inspection algorithm consists of two parts. One is a parameter learning algorithm, in which a genetic algorithm is used to extract optimal structuring element parameters, and segmentation and noise removal thresholds. The second part is a defect detection algorithm, in which the parameters obtained by a genetic algorithm are used for morphological operations. The machine vision system has been applied in an industrial setting to detect two types of casting defects: parts mix-up and any defects on the surface of castings. The system performs with a 99% or higher accuracy for both part mix-up and defect detection and is currently used in industry as part of normal production.

Genetic learning based texture surface inspection

This paper presents a novel approach of visual inspection for texture surface defects. It is based on the measure of texture energy acquired by a kind if high performance 2D detection mask, which is learned by genetic algorithms. Experimental results of texture defect inspection on textile images are presented to illustrate the merit and feasibility of the proposed method.

Anomalous evaporation behavior of ZnO powder milled mechanically under high-energy conditions

ZnO powder showed anomalous evaporation behavior after its mechanical milling treatment under high-energy conditions. The amount of generated vapor is about 10 times higher in the first 15 min of annealing at 1300 °C than that of unmilled ZnO powders. The strong ball impacts are responsible for the greatly enhanced evaporation ability. Low-energy ball milling involving shearing actions and rare weak impacts leads only to a small evaporation rate enhancement. The possible explanation of the high evaporation rate of the heavily milled material is the existence of large fraction of weakly bonded atoms in grain boundaries, surface defects and strained areas.

Hygrothermal effects on painted carbon fibre composite surfaces

In this article, the effect of hygrothermal aging on the painted surface finish of unidirectional and fabric carbon fibre composite laminates, with and without surfacing film was investigated. The results highlighted the importance of ensuring that the composite surface directly beneath the paint layer is made from a uniform material with a consistent thickness in order to minimise surface defects from occurring during aging. The surfacing film was found to minimise the print through development on the painted unidirectional and twill composite surfaces. However, the surfacing film layer was found to intermingle with the carbon fibre plies during cure, which resulted in an uneven film thickness that caused increased levels of orange peel. The twill laminate painted surface produced high levels of print through and surface waviness that was caused by the large resin rich regions located within the tow intersections at the surface which enlarged due to thermal expansion and swelling of the matrix with hygrothermal aging. It was also noted that the small resin rich regions between the individual carbon fibres on the unidirectional composite surface were sufficiently large to print through the painted surface.

Functionalisation of carbon fiber toward enhanced fiber-matrix adhesion

The surface of both oxidized and unoxidized unsized carbon fiber was functionalized using an aziridine linking group derived from reactive nitrenes, attempts were made to install pendant amines using amide chemistry. Surface functionalization using the nitrene approach was supported by X-ray Photoelectron Spectroscopy, in both oxidized and unoxidized carbon fiber. None of the chemical treatment pathways had a significant impact on the tensile strength of the individual fibers, and atomic force microscopy revealed that fibers undergoing these treatment methodologies remained intact, without creating additional surface defects.

Investigation of progress of reactions and evolution of radial heterogeneity in the initial stage of thermal stabilization of PAN precursor fibres

The relationship between process parameters and structural transformations in the fibres at each stage of the carbon fibre manufacturing process play a crucial role in developing high performance carbon fibres. Here we report a systematic method which uses the combination of Taguchi approach and scientific evaluation techniques to establish these relationships for the initial stage of thermal stabilization. Density, cyclization index and fraction of reacted nitriles of a precursor containing acrylonitrile, methacrylate and itaconic acid (AN/MA/IA) were used to assess the progress of stabilization in the fibres with respect to various combinations of process parameters. The extent of progress of stabilization improved with increase in temperature (from 225 to 235 °C) and time (from 12 to 24 min) whereas an opposite trend was observed with increase in the tension on the fibres from (1600-2550 cN). According to optical microscopy, radial heterogeneity was observed in the fibres treated at 235 °C. Interestingly, we were able to identify the existence of heterogeneous modulus distribution from skin to core of the precursor fibres which was further transferred to treated fibres. The overall radial modulus of treated fibres was higher than the precursor fibres. In contrast to the literature, the fracture morphology of the fibre samples indicated that initiation of crack is caused by surface defects rather than radial heterogeneity.

Corrosion behaviour of direct current and active screen plasma carburised AISI 316 stainless steel in boiling sulphuric acid solutions

Active screen plasma is a recently developed plasma surface alloying technique, which has shown potential for addressing some drawbacks associated with conventional direct current plasma processes. In this study, the corrosion performance of untreated, direct current and active screen plasma carburised AISI 316 was investigated by immersion in a boiling solution of sulphuric acid. The experimental results show that the corrosion behaviour of expanded austenite produced by low temperature plasma carburising is controlled by the type and density of surface defects; the corrosion properties of the active screen plasma carburised material are superior to that produced by direct current plasma because of the significantly reduced edge effect and surface defects; and the bias level used in the active screen carburising treatment has a profound effect on the corrosion performance of the material. Based on the experimental results, the corrosion mechanisms involved are discussed.

Study of active screen plasma processing conditions for carburising and nitriding austenitic stainless steel

Active screen (AS) is an advanced technology for plasma surface engineering, which offers some advantages over conventional direct current (DC) plasma treatments. Such surface defects and process instabilities as arcing, edge and hollow cathode effects can be minimised or completely eliminated by the AS technique, with consequent improvements in surface quality and material properties. However, the lack of information and thorough understanding of the process mechanisms generate scepticism in industrial practitioners. In this project, AISI 316 specimens were plasma carburised and plasma nitrided at low temperature in AS and DC furnaces, and the treated samples were comparatively analysed. Two diagnostic techniques were used to study the plasma: optical fibre assisted optical emission spectroscopy, and a planar electrostatic probe. Optimum windows of treatment conditions for AS plasma nitriding and AS plasma carburising of austenitic stainless steel were identified and some evidence was obtained on the working principles of AS furnaces. These include the sputtering of material from the cathodic mesh and its deposition on the worktable, the generation of additional active species, and the electrostatic confinement of the plasma within the operative volume of the furnace.

Formation of defects in the graphite oxidization process: a positron study

High-resolution positron annihilation lifetime (PAL) and two-detector coincidence Doppler broadening of annihilation radiation (2D-DBAR) measurements on graphite and its oxide derivatives for defect information, differing in oxidization agents, are reported. Positron measurements were found to be very effective in the investigation of defects in graphite and its derivatives. Positrons are mainly annihilated in vacancy-like defects on the particle surface and in large open-volume holes associated with the interface of graphite and graphite oxide. Different types of defects have been detected for unexfoliated graphite oxide and exfoliated graphene oxide based on 2D-DBAR measurements, namely the vacancy cluster and vacancy-oxygen complexes. It is also interesting to observe that the calculated large open-volume diameter of graphene oxide coincides with the distance between the layers from the XRD investigation, which indicates that the annihilation of the long-lived lifetime component τ3 might take place in the area between the graphene layers; no large open-volume hole has been detected.