- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
Non-destructive inspection of food and technical oils by terahertz spectroscopy
摘要： Quality control and non-destructive monitoring are of notable interest of food and pharmaceutical industries. It relies on the ability of non-invasive inspection which can be employed for manufacturing process control. We hereby apply terahertz (THz) time-domain spectroscopy as non-destructive technique to monitor pure and degraded oils as well as hydrocarbon chemicals. Significant differences in the spectra of refractive index (RI) and absorption coefficient arising from the presence of ester linkages in the edible and technical oils were obtained. Explicit increase from 1.38 to 1.5 of the RI in all THz spectrum range was observed in hydrocarbons and mono-functional esters with the increase of molar mass. This fact is in contrast of RI dependence on molar mass in multi-functional esters, such as Adipate or vegetable oils, where it is around 1.54. Degradation products, Oleic Acid (OA) and water in particular, lead only to some changes in absorption coefficient and RI spectra of vegetable oils. We demonstrate that complex colloidal and supramolecular processes, such as dynamics of inverse micelles and oil hydrolysis, take part during oil degradation and are responsible for non-uniform dependence of optical properties on extent of degradation.
关键词： absorption coefficient,terahertz spectroscopy,oils,degradation,non-destructive inspection,refractive index
更新于2025-09-23 15:23:52
-
An automated shearography system for cylindrical surface inspection
摘要： A shearography system has been developed to inspect the external heat proof coating bonded to a cylinder. The system consists of a shearography device integrated with thermal excitation, a mechanical translation and rotation device, and a central control unit. The translation and the rotation are driven with 2 servo motors. The combination of these two movements enables full inspection of the entire surface of the cylinder. The inspection sequence is automatically scheduled by inputting the geometry of the sample. Artificial intelligence (AI) has been first introduced to aid defect recognition from the resulted phase shifting fringe patterns. A recognition algorithm based on deep learning has been developed using Faster R-CNN model for recognition of bonding defects. By training the system using typical butterfly fringe patterns which are captured from bonding samples, the system can accurately identify the bonding defects on the cylindrical surface at a high success rate.
关键词： Defect recognition,Automatic inspection,Shearography,NDT
更新于2025-09-23 15:23:52
-
Study on the Statistical Errors in X-Ray Stress Measurement with Two-Dimensional Detector
摘要： The sin2 ψ method [1] is conventionally used well as how to measure non-destructively the residual strain and stress states of polycrystalline materials by X-ray diffraction. In the conventional method, there are D?lle-Hauk method [2] and Winholz-Cohen least squares analysis [3] as the determinations of the strain and stress states for limiting the influence of measurement errors. Many researches are made about the statistical error in those methods. In recent years, use of the X-ray stress measurements with two-dimensional detector from the conventional method is spreading. One of the measurements is called the cos α method. The measurement errors have attracted a great deal of attention for users as the spreads. Therefore, the basic equations and determinations of the strain and stress states are examined. The confidence intervals of measured stress by the cos α method. The research and development is performed for the the cos α method which took the influence of measurement errors into consideration.
关键词： Cos α method,Non-destructive inspection,Statistical errors,Two-dimensional detector,X-ray stress measurement,internal stress
更新于2025-09-23 15:23:52
-
[IEEE 2018 International Conference on Cyberworlds (CW) - Singapore, Singapore (2018.10.3-2018.10.5)] 2018 International Conference on Cyberworlds (CW) - Towards Automatic Optical Inspection of Soldering Defects
摘要： This paper proposes a method for automatic image-based classification of solder joint defects in the context of Automatic Optical Inspection (AOI) of Printed Circuit Boards (PCBs). Machine learning-based approaches are frequently used for image-based inspection. However, a main challenge is to manually create sufficiently large labeled training databases to allow for high accuracy of defect detection. Creating such large training databases is time-consuming, expensive, and often unfeasible in industrial production settings. In order to address this problem, an active learning framework is proposed which starts with only a small labeled subset of training data. The labeled dataset is then enlarged step-by-step by combining K-means clustering with active user input to provide representative samples for the training of an SVM classifier. Evaluations on two databases with insufficient and shifting solder joints samples have shown that the proposed method achieved high accuracy while requiring only minimal user input. The results also demonstrated that the proposed method outperforms random and representative sampling by ~ 3.2% and ~ 2.7%, respectively, and it outperforms the uncertainty sampling method by ~ 0.5%.
关键词： Classification of solder joint defects,active learning,Automatic Optical Inspection (AOI),SVM classifier,K-means
更新于2025-09-23 15:23:52
-
[IEEE 2018 IEEE International Conference on Imaging Systems and Techniques (IST) - Krakow, Poland (2018.10.16-2018.10.18)] 2018 IEEE International Conference on Imaging Systems and Techniques (IST) - Intelligent Navigation and Control of a Prototype Autonomous Underwater Vehicle for Automated Inspection of Aquaculture net pen cages
摘要： Aquaculture is one of the fastest growing food sectors worldwide providing more than 50% of world fish consumption. Towards sustainable development, aquaculture ought to design and implement technical solutions for the efficient management of farms, thus improving fish performance and decreasing operational costs, human effort and environmental impact. For this, periodic inspection of fish-cages along with early warning systems are required, functionalities that small-sized underwater vehicles can provide. In this paper, an efficient methodology for intelligent navigation of an Autonomous Underwater Vehicle (AUV) architecture is presented, yielding to a useful tool with advanced capabilities in terms of automated manipulation via real-time optical recognition approaches, miniaturization of sensors and processing units, selective monitoring, data recording/transmission operations and proper parameter calculations for further offline-analysis of captured information. The proposed AUV system constitutes an increased Technology Readiness Level version of the preliminary prototype designed earlier by our group, incorporating additional modules and application capabilities, focusing on regular periodic fish-cage net inspection in terms of net holes and fouling. The optical navigation scheme has been tested in laboratory installations under numerous scenarios so as to determine the factors affecting its robustness and efficiency. Results extracted under the validation procedure in operational conditions indicate that the proposed framework can prove a cost-effective, flexible and operative solution for aquaculture industry, enabling the transfer of operations further offshore.
关键词： remote operation,automated navigation,optical recognition,machine vision,net inspection,autonomous underwater vehicles
更新于2025-09-23 15:22:29
-
A monocular vision–based perception approach for unmanned aerial vehicle close proximity transmission tower inspection
摘要： Employing unmanned aerial vehicles to conduct close proximity inspection of transmission tower is becoming increasingly common. This article aims to solve the two key problems of close proximity navigation—localizing tower and simultaneously estimating the unmanned aerial vehicle positions. To this end, we propose a novel monocular vision–based environmental perception approach and implement it in a hierarchical embedded unmanned aerial vehicle system. The proposed framework comprises tower localization and an improved point–line-based simultaneous localization and mapping framework consisting of feature matching, frame tracking, local mapping, loop closure, and nonlinear optimization. To enhance frame association, the prominent line feature of tower is heuristically extracted and matched followed by the intersections of lines are processed as the point feature. Then, the bundle adjustment optimization leverages the intersections of lines and the point-to-line distance to improve the accuracy of unmanned aerial vehicle localization. For tower localization, a transmission tower data set is created and a concise deep learning-based neural network is designed to perform real-time and accurate tower detection. Then, it is in combination with a keyframe-based semi-dense mapping to locate the tower with a clear line-shaped structure in 3-D space. Additionally, two reasonable paths are planned for the refined inspection. In experiments, the whole unmanned aerial vehicle system developed on Robot Operating System framework is evaluated along the paths both in a synthetic scene and in a real-world inspection environment. The final results show that the accuracy of unmanned aerial vehicle localization is improved, and the tower reconstruction is fast and clear. Based on our approach, the safe and autonomous unmanned aerial vehicle close proximity inspection of transmission tower can be realized.
关键词： Close proximity inspection of transmission tower,monocular vision,UAV self-positioning,tower localization
更新于2025-09-23 15:22:29
-
Defect detection techniques robust to process variation in semiconductor inspection
摘要： As semiconductor manufacturing process has resulted in downscaling device dimensions, the critical defect size has been becoming smaller and smaller. The highly sensitive optical wafer inspection tool for detecting small defects erroneously detects the process variations as defects and generates a large amount of 'nuisance' information. Thus, the scanning electron microscope (SEM)-based review tool needs to automatically discriminate between defects and nuisance information. To discriminate nuisance information, the absence of defects in the SEM image needs to be accurately detected through an inspection process using the review tool. We propose a defect detection method with (a) an integration of multiple comparison-detection results (IMCD) to suppress the number of defect candidates and (b) a discrimination based on a normal image model (DNPM) to judge whether the candidate is a defect or normal. An evaluation using SEM images of a processed wafer revealed that combining the IMCD and DNPM achieves a nuisance information discrimination rate of 84.4% and a defect detection rate of 93.3%, which are higher than those of the one-class support vector machine (SVM). The proposed methods automatically collect defect images efficiently even when much nuisance information is produced by the optical wafer inspection tool and enable manual visual checks to be reduced.
关键词： image processing,clustering,one-class discrimination,self-organizing map,defect inspection
更新于2025-09-23 15:22:29
-
Virtual subpixel approach for single-mask phase-contrast imaging using Timepix3
摘要： X-ray phase contrast imaging provides a method to distinguish materials with similar density and effective atomic number, which otherwise would be difficult using conventional X-ray absorption contrast. In recent years, multiple methods have been developed to acquire X-ray phase contrast images using incoherent laboratory sources. The single mask edge illumination setup has been demonstrated as a possible candidate for large scale applications due to its relaxed restrictions on longitudinal coherence and mask alignment, and for its ability to do bi-directional phase contrast images in a single sample exposure. Unfortunately, the single mask edge illumination setup's refraction sensitivity, and thereby signal to noise, is limited by detector artifacts. Furthermore, it requires multiple exposures to perform dark-field imaging, a method that enables imaging of micro-structures smaller than the image resolution. We propose using an Advapix detector with Timepix3 pixel-readout chip in a single mask imaging setup to improve signal to noise ratio in phase contrast images. This is achieved using the Timepix3 chip's ability to simultaneously acquire fast time of arrival and time over threshold measurement of single photon events, which enables sub-pixel identification of individual photons. In this paper, we demonstrate that signal to noise ratio can be improved by at least 67 ± 5 % using subpixel identification of single photons compared to conventional acquisitions methods. Thereby the required sample dose can be reduced considerably. This shows that there is a great potential in using Timepix3 chip to improve x-ray phase contrast imaging. Further, the results indicate the possibility for dark field imaging in a single sample exposure using Timepix3 in a single mask edge illumination setup.
关键词： X-ray detectors,X-ray radiography and digital radiography (DR),Data processing methods,Inspection with x-rays
更新于2025-09-23 15:22:29
-
[IEEE 2018 31st International Vacuum Nanoelectronics Conference (IVNC) - Kyoto, Japan (2018.7.9-2018.7.13)] 2018 31st International Vacuum Nanoelectronics Conference (IVNC) - Silicon Cold Field Emission Electron Sources for Electron Microscopy Inspection Systems
摘要： Next generation electron beam inspection tools will require a high brightness electron source. We have designed, microfabricated and tested arrays of metal coated silicon cold field emitters that have high current density and high reduced brightness, offering a promising alternative to thermal field emitters. One of the main challenges that must be addressed with cold field emitters is achieving stable emission current. Using current pulsing, we have been able to improve the stability of field emission currents and have demonstrated beam currents with less than 1% noise.
关键词： electron source,microfabrication,cold field emission,electron microscopy,inspection systems,cathode,silicon
更新于2025-09-23 15:21:21
-
[IEEE 2019 IEEE 46th Photovoltaic Specialists Conference (PVSC) - Chicago, IL, USA (2019.6.16-2019.6.21)] 2019 IEEE 46th Photovoltaic Specialists Conference (PVSC) - Aerial Infrared Thermography of a CdTe Utility-Scale PV Power Plant
摘要： Aerial Infrared Thermography (aIRT) is a fast and flexible inspection method to monitor and assess utility-scale photovoltaic (PV) power plants. The literature is abundant on aIRT for crystalline silicon (c-Si) modules, but very little investigation has been carried out and reported thin-film PV. As Cadmium Telluride (CdTe) is currently the leading thin-film technology, and a good performer in warm and sunny climates, this paper aims to investigate the application of aIRT on CdTe PV plants in Brazil. Results demonstrate that aIRT is a reliable, cost-effective and fast method to detect faults on CdTe modules in large-scale PV plants.
关键词： Fault Inspection,Unmanned Aerial Vehicles (UAV),Photovoltaic Power Plants,Aerial Infrared Thermography (aIRT),Thin-Film,Cadmium Telluride (CdTe)
更新于2025-09-23 15:21:01