Papers by Keyword: Visual Inspection

Abstract: We have developed methodologies for the detection of microdefects on painted surfaces and the evaluation of machined surface quality through the use of the patterned area illumination. The objective of this study is to optimize illumination for surface inspection of products by reproducing the illumination situation using the patterned area illumination. In this study, a ray tracing method, which is a computer graphics technique, was employed to reproduce a scenario in which a product surface is illuminated by the aforementioned patterned area illumination. In the conventional ray tracing method, the patterned area illumination is not taken into account. Accordingly, the surface to be inspected was postulated to be a mirror surface, and the illumination pattern was projected onto it in order to reproduce the reflection of the illumination pattern on the inspection surface. The developed simulation method enables several key optimizations. Firstly, it allows for the optimization of the illumination device by reproducing the inspection surface with a curved surface. Secondly, it enables the data augmentation of teacher data for machine learning for a versatile defect detector. Thirdly, it allows for the optimization of the circular pattern used to estimate the shape of micro defects. Finally, it enables the reproduction of pattern projection onto a machined surface for the evaluation of machining quality of the machined surface.

Abstract: Friction stir welding (FSW) can be classified as a new non-fusion welding technique if compared to other conventional welding techniques. In this study, a thin Zn-coated metal sheet was welded using the FSW technique with a lap-joint configuration. The weld joint was inspected using non-destructive testing (NDT) methods to determine the quality of the weld produced. There are two main parameters used in the experiment, which are rotational speed [RPM] and traverse [mm/min]. The design of the welding experiment is carried out using the surface response technique with the circumscribe central composite design method. This experiment was conducted using two factors and two levels of design of the experiment setup. All the samples are successfully welded and examined by visual inspection and radiographic testing (RT) to identify defects in the weld. Welding defects are plotted into the graph to assist in the process of analyzing the trend of defects. Several defects have been discovered such as flash formation, thinning material, through-hole, and high-density solid inclusion during the experiment. As a result of the study, it is possible to make a preliminary overview of the appropriate parameters used for FSW of thin Zn-coated metal sheets. In addition, welding defect information from the NDT inspection can assist in the process of preparing specimens for destructive tests.

Abstract: This study presents the results of visual and infrared (IR) inspection of photovoltaics (PV) technologies installed at the Qatar environment and energy research institute (QEERI) outdoor test facility (OTF) at Qatar Foundation (Doha, Qatar). Silicon based PV technologies which have been operational in the field since 2014, have been investigated for various failure modes. The visual inspections were carried out for all the PV modules from the backside however, the inspection from the front side was not possible for some modules due to heavy soiling. The visual defects which were identified during this study include, cracking of the back glass, yellowing of the encapsulant material, cracks formation in the back sheet, and pits formation in the back sheet. The visual inspection revealed that around 19 % of the total modules have back sheet cracking and discoloration, 8 % have yellowing of the encapsulant, and around 4 % were having pits in the back sheet. Moreover, one module was detected with back glass cracking. The IR inspection was also done both from front and backside for all the silicon PV modules to detect hot spots. The IR inspection has revealed that hot spots were generated at different locations of the PV modules. 39 % of the modules have hot spots at the location of junction boxes, around 6 % of the modules have hot spots in junction boxes and around 1 % have hot spots at the locations away from junction boxes. The visual and IR inspection has revealed that the dominant failure modes which have been observed for silicon-based technologies at OTF are the hot spots generation at junction boxes and the back sheet cracking, and its yellowing.

Abstract: Building inspection is the first step before building maintenance. Without building inspections, maintenance work will be overwhelming and will result in losses in terms of time and money incurred. To determine the defect, visual inspection will be conducted as early phase of building inspection. In Malaysia, inspection on exterior enclosure of structures not usually apply because of the high cost of doing inspection and it takes a long time too. Furthermore, visual inspection usually carries out with the aid of binoculars and cameras for examining eyes. In general, the buildings that known as a high-rise building is a building that has the structure roughly seven floor or more. At such altitude, any defects and damage to the building difficult to be observe with the naked eye. Hence, the aim of this study is to propose Unmanned Aerial Vehicle (UAV) as tool that used to inspecting defect on high-rise building. Building that being choose is Urban Transformation Centre (UTC) Malacca that located at Jalan Hang Tuah, 75300 Malacca. The data collected will be analyzed using Condition Survey Protocol (CSP1 Matrix) to obtain an overall rating of the building’s condition. The process involved in this study are site visit, planning and preparation before the flight, data collection using UAV, image processing and analyzing data from image. The results of analysis, 331 defects have been identified and documented suffer from defects or faults such as cracking, scaling, spalling, peeling paint, expansion joint and mould. All defects information obtained using UAV visual inspection can help the UTC Malacca maintenance to do maintenance work. The overall condition of this building structure is 6.6, where the overall performance of the building is average.

Abstract: The final set of properties of the exhaust valve is formed in the course of the entire set of processes, the most important is the build-up of the valve chamfer which is subjected to constant high temperatures, gas flow and thermal cycling, therefore strict adherence to the whole complex of properties is necessary. In the study two materials were investigated: the deformable part of the base material (steel grade 40X10C2M (EN steel grade X40CrSiMo10)) and surfacing (alloy EP616A), used to manufacture the exhaust valve of the internal-combustion engine.The study was carried out at different stages of valve manufacturing: after electro-forging during the formation of a valve head from a rod, subsequent heat treatment (quenching + high tempering), after surfacing a ring with the heat-resistant alloy. The chemical composition of phases, inclusions, and sections of EP616A alloy around the shrink-hole were determined by the method of micro-X-ray analysis. The results indicate a shortage of visual inspection, and that conducting a selective destructive method does not ensure the full probability of detecting internal defects. It is advisable instead of destructive control to introduce ultrasonic flaw detection.

Abstract: This work is based on making a comparison of different inspection methods of non-destructive testing (NDT), to detect porosity, bite, undercut, splash, overlap, slag, concavity, lack of fusion and damage to base material, in welds of shielded metal arc welding (SMAW) and flux-cored arc welding (FCAW), made on a steel bridge ASTM-A-588. The main application of non-destructive tests is to evaluate the final state of a welded joint in critical points, in addition the fundamental objective is to ensure that the established acceptance and rejection criteria are met rigorously. For which non-destructive techniques are analyzed, such as the visual testing of welding on the beams of a bridge of length 35000 mm, subsequently, tests of inspection by liquid penetrant were used, based on the AWS D1.5 standard to confirm the failures in welded joints in ABA webs, an ultrasonic is also carried out using an SIUI CTS 602 equipment, finally to issue an acceptance and rejection criterion as determined by ASTM E165 and ASTM E2544-09.

Abstract: For precise assembly of miniature parts, the precise inspection for parts’ posture and real-time servo control for assembly greatly depend on the performance of visual inspection system. This paper proposed a high-resolution real-time visual inspection system of micro assembly. The CMOS image sensor and high-speed digital signal processing chip were chosen to design the image acquisition module, image processing module and image display module. High-accuracy display on the common display device was implemented with the video encoding chip and FPGA. The test results showed that the processing speed with preprocessing could reach 3.5 frames per second with 5 mega-pixel resolution, and the display accuracy after threshold processing had little loss. Micro parts assembly experiment and high accuracy Peg-in-Hole assembly experiment are done to test the performance of the proposed visual inspection system. This visual inspection system can be used for high-resolution real-time micro assembly and other real-time visual servo control.

Abstract: This study discussed abrasive machining of aspheric glass lens, with the aim to increase the aspheric lens machining efficiency and reduce the back-end polishing process. The discussions covered "optical lens design", "Taguchi method optimization", "grinding force measurement", and "on-line real-time inspection". A window interface was designed by using Borland C⁺⁺ Builder for optical glass design. The aspheric design parameters were substituted in it to obtain the machining path. In terms of Taguchi method optimization, the performance of electroformed grinding rod was compared with that of resin grinding rod. Taguchi method was used to look for optimization, and the quality characteristic was surface roughness. After experimental analysis, the optimization results were: using a resin binder grinding rod, the optimum parameter of 50,000 (rpm), forward grinding, feeding at 10 mm/min, grinding depth at 0.02mm, and aspheric lens surface roughness of 0.03μm. In terms of grinding force measurement, an eight-ring strain gage dynamometer was designed to measure the optimum grinding force. The measurement result showed that the grinding force in vertical direction was 3.2kg, and that in horizontal direction is 4kg. In terms of visual inspection, a CCD camera was mounted on the machine, which takes the image of workpiece immediately after it is machined. The form was inspected, and the dismounting error was reduced.

Abstract: To reduce error inspection rate of the brand inspection in enterprise logistics, this paper puts forward a inspection method which depends on template matching algorithm based on machine vision.This method identifies cigarette carton brand by image processing, first template matching, accurate template matching of real-time carton bar code, then sends message to industrial personal computer to does sorting operation. Practice has proved that the system is stable, fast, accurate to meet the site requirement.

Abstract: This article will use visual inspection techniques into the two-dimensional geometry measurement of ceramic products .It uses MATLAB 7.0 as image processing software, and the designing uses the gradient information into detection circular edge by Hough transformation method. The method is helpful to improve the production efficiency and the recognition rate. We can prove that this method has the feasibility by experiments, and achieve a satisfactory result .