Glass Defects Research Articles

X-ray-induced long persistent luminescence of Cu+-doped high alumina borosilicate glass

A Cu+ doped high alumina borosilicate glass with long persistent luminescence (LPL) was prepared by reduction melting and radiation induction based on glass defect engineering. The maximum time of the persistent luminescence can arrive 16 h. Spectral and EPR measurements indicate that two kinds of oxygen defects can be induced by X-ray radiation. One defect was eliminated after UV irradiation, and the other can synergistically participate in the LPL process of glass with reduction induced oxygen deficiency centers(ODCs). In addition, there is a highly linear relationship between the initial LPL intensity of the glass and the radiation dose; which implies the glass can be used as a radiation detection material. The internal mechanism of the linear relationship and LPL was discussed.

Glass Defect Detection with Improved Data Augmentation under Total Reflection Lighting

To address the technical challenge of identifying tiny defects, especially dust and point defects, on mobile phone flat glass, an automatic optical inspection system is established. The system investigates algorithms including imaging principles, target detection models, data augmentation, foreground segmentation, and image fusion. The system builds an automatic optical inspection platform to collect glass defect samples. It illuminates the glass samples with a combined total reflection–grazing light source, collects the defect sample data, segments the background and defects of the collected data, generates the defect mask, and extracts the complete defects of the cell phone flat glass. The system then seamlessly integrates the extracted defects with a flawless background using Poisson editing and outputs the location information of the defects and the label output to automatically generate the dataset. The deep learning network YOLOv5 works as the core algorithm framework, into which the Constructive Block Attention Module and the small target detection layer are specifically added to enhance the capability of the model to detect small defects. According to the experimental results, the combined lighting effectively improves the precision of detecting dust and bright spots. Additionally, with the adoption of novel data augmentation techniques, the enhanced YOLOv5 model is capable of effectively addressing the challenges posed by inefficient sample data and non-uniform distribution, thus mitigating network generalization issues. Furthermore, this data augmentation approach facilitates the rapid adaptation of the same detection tasks to diverse environmental scenarios, enabling the expedited and efficient deployment of the model across various industrial settings. The mean average precision (MAP) of the optimal model in the validation set reached 98.36%, 2.62% higher than that of the original YOLOv5. In addition, its false acceptance rate (FAR) is 1.27%, its false rejection rate (FRR) was 2.47%, its detection speed was 64 fps, and its correct detection rate in the validation set was 98.75%, which meets the current industrial detection requirements by and large. In this way, this paper achieved the automated inspection of mobile phone flat glass with high robustness, high precision, and a low false acceptance rate and false rejection rate, significantly reducing material losses in the factories and the likelihood of error occurrence in follow-on products. This method can be applied to the multi-scale and multi-type detection of glass defects.

Identifying Alternative Techniques for the Characterization of Glass Defects

Cam üretiminde ortaya çıkan hataların karakterizasyonu, cam üretim sürecini etkileyen en önemli faktördür. Doğru ve hızlı karakterizasyon, hata kaynağının tespiti ve hataların oluşumunun engellenmesi bakımından önem teşkil etmektedir. Cam hatalarının katı ya da gaz, amorf (düğme/damar) ya da kristalin (taş) olması hatanın kaynağını bulmakta uygulanacak tekniğin belirlenmesi için önemlidir. Farklı katı cam hatalarının özelliklerinin belirlenmesinde farklı analiz teknikleri uygulanmıştır. Cam kalitesini etkileyen katı cam hatalarının (taş) karakterize edilerek tanımlanmasını ve hata kaynağının doğru tespit edilmesini amaçlayan bu çalışmada; cam hataları mevcutta kullanılmakta olan Taramalı Elektron Mikroskobu ve Enerji Dağılımlı X-ışını Spektroskopisi yöntemine alternatif olarak Mikro X-ışını Kırınımı, Mikro Raman Spektroskopisi ve Fourier Dönüşümlü Kızılötesi Spektrometresi yöntemleri ile incelenmiştir. Taramalı Elektron Mikroskobu ve Enerji Dağılımlı X-ışını Spektroskopisi yöntemiyle cam hatalarının karakterizasyonu, mikroyapı incelemesi ve yarı kantitatif kimyasal analizi ile birlikte yapılmaktadır. Mikro X-ışını Kırınımı ile hata, faz tespiti ile kalitatif bir şekilde tanımlanmaktadır. Mikro Raman Spektroskopisi ve Fourier Dönüşümlü Kızılötesi Spektrometresi ise kristalin inklüzyonların (taş hatası) içerdiği moleküler bağ karakterizasyonuna göre sonuç vermektedir. Çalışmada farklı türde hataların incelemeleri bu metotlarla yapılarak bu dört yöntemin karşılaştırması yapılmış ve bunların uygulanabilirliği değerlendirilmiştir. İncelenen dört yöntemin farklı temellere dayanan analiz tekniği olmasından dolayı çıkan sonuçların yorumlanması oldukça önemlidir. Taramalı Elektron Mikroskobu ve Enerji Dağılımlı X-ışını Spektroskopisi yönteminde mikroyapı incelemesi ve yarı kantitatif sonuçlar değerlendirilirken, Mikro X-ışını Kırınımı yöntemiyle yapılan analizlerden faz tayini ile kalitatif tespit yapılabildiğinden katı cam hatalarının oluşum sıcaklıkları hakkında da fikir elde edilebilmektedir. Mikro Raman analizlerinde de kalitatif sonuçlar alınırken hatanın homojen olması gerektiği görülmüştür. Fourier Dönüşümlü Kızılötesi Spektrometresi analizlerinde alınan spektrum ile kalitatif sonuçlara ulaşılsa da verilerin yorumlanması için detaylı bir literatür araştırması gerekmektedir. Hata tayininde en uygun yöntem hata türüne ve kristal yapısına bağlı değişkenlik göstermekle birlikte, stereo mikroskop ve SEM-EDS ile edinilen bulguları faz tayiniyle destekleyen µXRD olacaktır. Faz tayininin önemi, kristalin oluşum sıcaklığı konusunda yorum yapılabildiğinden hatanın kaynağı konusunda öngörü yapma olanağı sunmasıdır.

Reduction of injection molded silica glass defects and enhancement of glass quality via water debinding

Pure silica glass has been fabricated by the injection molding and sintering technique. The quality of the as-fabricated silica glass was improved via the water debinding process.

Experimental repair technique for glass defects of glass-glass photovoltaic modules – A techno-economic analysis

Solar photovoltaic (PV) energy is a crucial supply technology in the envisioned renewable energy system. With enormous amounts of PV modules being installed, some will be affected by early-life failures and the resulting e-waste from PV modules is raising environmental concerns. A failure of growing importance is the defect in the glass layer(s) of glass-glass PV modules. In this research, an experimental glass repair technique for glass-glass PV modules was tested and examined. The PV modules with glass defects under test did not show internal defects in the PV cells, while the repaired specimens performed properly at each phase in the repair process compared to reference modules, the IEC standards and manufacturer warranty. After a damp-heat test the repaired PV modules showed no signs of water ingress, suggesting that the glass layer was restored as a proper barrier. However, definite conclusions should be made with caution since the non-repaired specimens neither showed visible signs of water ingress. While the practical application of the reparation technique has still some uncertainties, glass reparation is found to be technically feasible and effective. Furthermore, economic and energetic analyses indicate that glass defect reparation is economically interesting and energetically desirable.

Machine vision-based cutting process for LCD glass defect detection system

In this research, the automatic optical detection system is developed for detecting the sectional profile and the surface of the thin-film transistor liquid crystal display (TFT-LCD) panels after being treated through the cutting process. Traditional image processing inspection relying on pre-determined thresholding cannot achieve ideal results in slight defects in glass substrates. The proposed image pre-processing process was integrated with the deep learning technique to further enhance the detection process of inconspicuous defects in glass substrates. In addition, the photoelastic reflection lighting technique was used to highlight subtle defects in low-contrast surface images of glass substrates. When tested with the sectional profile photodetector, uniformed lighting effect is achieved by combining the concentrated light source of the inner coaxial lens with the line light source in order to test the surface coarseness-related characteristics of the glass sectional profile so as to indicate the defect while intensifying the contrast effect of the sectional profile background. When detecting the sectional profile features, it is conducted by separating the rib mark features through the U-Net network model developed for the deep learning; in result, 100% accuracy can be achieved. When detecting the sectional profile defect, Auto Encoder network model of the deep learning is used to learn the background picture retrieved from the original picture through the linear regression process. As a next step, the model is used again to predict the result by subtracting the background picture from the original picture, and then the defect position is highlighted; as a result, 98% accuracy is achieved. When detecting the model acceleration, it is conducted by revising the model weighting data format. In terms of U-Net Model, the reading time has been shortened for 4.28 s; in individual picture, the prediction time has been shortened for 0.29 s; in Auto Encoder, the model reading time has been shortened for 19.23 s; and the individual picture prediction time has been shortened for 0.94 s. As for the surface detection, the circular polariscope is developed. During the detection, the photoelastic reflection theory is employed by projecting the circularly polarized light vertically onto the panel surface in order to produce the interfering halo at the deformed area surrounding the denting defect, and the resulting features are referenced to identify the denting defect. By screening the mean luminance value of the sliding window and the discrete value, 90% detection accuracy can be achieved. In the meantime, it can also be used to pinpoint the denting defect that is over 3° in average angle change as seen on the normal line surrounding the defect.

Multi-scale based defect detection for automotive glass

Small targets have low resolution and small size. Using the YOLOv3 model may not detect the target object. In view of this situation, this paper improves the YOLOv3 model and proposes a network structure with refined features. The channel feature refinement mechanism is introduced, and four detection scales are designed to prevent small objects from being submerged; and the spatial pyramid structure-Spp module is added to detect five defects of glass. The experimental results show that compared with the YOLOv3 model, the mAP value of the optimized model is 97.22%, and the mAP value of the original version is 92.31%, an increase of 4.91%. In the self-built data set, the accuracy of automotive glass defect recognition reached 97.66%, and the optimized model’s recognition effect on small targets was significantly improved. Compared with the traditional glass defect detection method, the method improves the detection accuracy and the recognition accuracy.

Deep Learning Application in Detecting Glass Defects with Color Space Conversion and Adaptive Histogram Equalization

Manually detecting defects on the surfaces of glass products is a slow and time-consuming process in the quality control process, so computer-aided systems, including image processing and machine learning techniques are used to overcome this problem. In this study, scratch and bubble defects of the jar, photographed in the studio with a white matte background and a -60° peak angle, are investigated with the Yolo-V3 deep learning technique. Obtained performance is 94.65% for the raw data. Color space conversion (CSC) techniques, HSV and CIE-Lab Luv, are applied to the resulting images. V channels select for preprocessing. While the HSV method decreases the performance, an increase has been observed in the CIE-Lab Luv method. With the CIE-Lab Luv method, to which is applied the adaptive histogram equalization, the maximum recall, precision, and F1-score reach above 97%. Also, Yolo-V3 compared with the Faster R-CNN, it is observed that Yolo-V3 gave better results in all analyzes, and the highest overall accuracy is achieved in both methods when adaptive histogram equalization is applied to CIE-Lab Luv.

Glass Defect Detection via Multi-Scale Feature Fusion

Glass defect detection is significant in glass industry. However, most of the existing methods for glass defect detection currently still rely on manual screening with high-cost and poor-efficiency. To address this issue, we propose a glass defect detection method using multi-scale feature fusion strategy. Specifically, we first propose an algorithm based on pix2pix to realize the edge extraction. Then, we propose a glass defect detector based on both local and global features. Comprehensive experiments are conducted on collected glass dataset from factories. The experimental results demonstrate that our method outperforms conventional methods including the traditional image processing and Holistically-Nested Edge Detection (HED), with the precision rate (PR) up to 97%, the false precision rate (FP) below 2% and the total accuracy rate (ACC) of glass defect detection up to 98%.

Benefits of optical distortion measurement – How Moiré technology drives the efficiency of glass production chains

High‐quality glass is facing a significantly growing demand. Especially display glass and automotive applications are boosting the market and the industrial pursuit for highest quality, requiring not only defect free sheets but also flawless optical properties. The technology to be introduced in this paper concerns the measurement of optical power inhomogeneities of glass, additional to reliably detecting cords, reams, zip lines and other common optical glass defects. Quality monitoring also comprehensively gives hints for the optimization of the entire float glass process via product and quality data. By relying on the ISRA technology, manufacturers can ensure the delivery of defect‐free float glass with certified optical quality regarding Diopter or Zebra values. The paper provides a closer look at ISRA's patented innovative optical distortion measurement, giving an overview about how users can benefit from the in‐line Moiré technology, which ensures highest product quality and enhances global competitiveness.

Annotation-free defect detection for glasses based on convolutional auto-encoder with skip connections

Glasses are fundamental materials for modern society. Therefore, strict quality control is significant for glass manufacturing. The current defect detection methods for glasses rely heavily on manual annotations. However, in some occasions, it is hard to acquire the manual annotations. To solve this problem, an unsupervised learning method, namely a convolutional auto-encoder with skip connections (SC-CAENet), is proposed in this paper. Firstly, a convolutional auto-encoder is applied to reconstruct the normal images. Afterwards, image difference is performed to detect the defect regions. In addition, skip connections and weighted structural similarity loss are introduced to improve the detection precision. The results show that the accuracy, escape rate, overkill rate, precision, recall and f1-score of SC-CAENet reach 0.9475, 0.025, 0.08, 0.924, 0.975 and 0.949, respectively. Moreover, the SC-CAENet can detect an image in 32 ms, which indicates that the SC-CAENet can precisely and efficiently detect glass defects.

Effect of melting atmospheres on the optical property of radiation-hard fluorophosphate glass

High-energy radiation in space and nuclear irradiation environment induces colour centres in optical glass, causing solarisation, and a serious condition can render optical systems and optical loads unusable. To develop space radiation-resistant optical glass, CeO2-stabilised radiation-hard fluorophosphate glass was prepared under three different atmospheres (nitrogen, oxygen, and ambient air). The glass-melting atmospheres' effects on the glass's transmission, defect formation, and structural changes before and after exposure to gamma radiation were investigated by a comprehensive study on their transmittance, absorption, and electron paramagnetic resonance spectra. Introducing a small amount of CeO2 (～0.34 wt%) into the fluorophosphate base glass converted NBO and BO into ABO in the glass network, red-shifted the UV absorption edge, and decreased the optical density increment by almost half after radiation. As the total dose of gamma radiation increased, the transmittance of the irradiated glass at a wavelength of 385 nm significantly increased due to absorption of POHC2 defects. After exposure to 250 k of rad gamma irradiation, the corresponding optical density increment per centimeter thickness at 385 nm of the radiation-hard fluorophosphate glass that melted in the nitrogen, oxygen, and air atmospheres decreased from 1.839 to 1.388 and 1.215. As it melted in air, the NBO ratio of the fluorophosphate glass reached the lowest level and the Ce4+ ratio in the glass was 92.49%, which helped suppress the generation of POHC, Fe3+, PO4-EC, and PO3-EC defects during the gamma irradiation process, improving the glass's radiation resistance.

Using glass defect engineering to obtain order–disorder transformation in cathode for high specific capacity lithium ion battery

Vanadium-based glass shows high electronic conductivity without grain boundaries. It easily undergoes structural microcrack defects during fabrication using the melt-quenching method resulting in voids that affect its properties. This paper demonstrates that the cathode material obtained by nanocrystals precipitated by long-cycle electric fields are identical with those obtained by the heat treatment method. V2O5-Li3PO4 (VP) glass precipitates nanocrystals of Li3P and Li0.3V2O5 under heat treatment at 340 °C. It showed a prominent specific capacity of 269.4 mAh.g−1 in the first cycle at a current density of 50 mA.g−1, and 227.4 mAh.g−1 after 50 cycles with a retention rate of 86.3%, as the cathode of a lithium ion battery. Thermal field-controlled VP glass allows selective crystallization of nanocrystal of Li3P and Li0.3V2O5, which promote conductivity and charge transfer and significantly enhance the electrode reaction kinetics. Through ex-situ TEM, DSC, XRD and XPS, it was observed that discharge/charge caused disordered transitions in the amorphous VP, leading to the formation of nanocrystals of Li3P and Li0.3V2O5 after 100 cycles. Amorphous glass electrode material as a high-energy unstable state, has same tendency of precipitation of crystals to become stable under different treatment methods with long electric field cycles and thermal field induction.

Manufacturing‐induced surface contaminations

SummaryIn the course of classical optics manufacturing glass components are in permanent direct contact with aqueous operating materials. Such contact leads to a certain absorption of water and hydrogenous compounds that may induce severe glass defects. In this contribution, absorption of hydrogen and other contaminants during grinding of glass was observed and qualitatively detected via laser‐induced breakdown spectroscopy. It is shown that hydrogen, calcium, magnesium, and carbon are implanted into the glass surface where the contaminant concentration increases over grinding time or contact time of the glass surface with water and the grinding tool, respectively. The contaminants hydrogen, calcium, and magnesium can be attributed to the water used as lubricant. In contrast, carbon most likely originates from wear debris of the used silicon carbide grinding pads. Several possible mechanisms that lead to such surface contamination of glasses during grinding – diffusion, accumulation in micro cracks as well as the formation of hydrated silica – are finally introduced.

Theoretical and experimental gamma-rays attenuation characteristics of waste soda-lime glass doped with La2O3 and Gd2O3

The present study explored the gamma-rays shielding characteristics of waste soda-lime glass (SLG) containing lanthanum oxide (La2O3) and gadolinium oxide (Gd2O3) by utilizing theoretical and experimental determinations. For this purpose, both La2O3, and Gd2O3 were separately added as 0.005, 0.05, and 0.5 wt% in place of waste SLG. The glass series of SLGL and SLGG were successfully fabricated via conventional melting techniques. Some physical analyses were done while radiation shielding parameters were figured out by using gamma-rays spectroscopy and XCOM computations to the produced glass specimens. One can easily state that the synthesized glasses showed a transparent appearance without any glass defects. Further, the glass density revealed an increasing trend with the insertion of both substances, but the SLGG series enhanced more. According to the experimental measurements, it was figured out that both oxide additions had the ability to increase the linear attenuation coefficient (LAC) values. The LAC was improved from 0.151 to 0.242 cm−1 for Gd2O3 insertion ration whereas 0.148 to 0.202 cm-1 for La2O3 series. Moreover, a good agreement between the experimental measurements and theoretical calculations was fulfilled. Based on the measured LAC values, other important parameters including mass attenuation coefficient (MAC), half-value layer (HVL), and mean free path (MFP) were evaluated for the investigated glass series. All in all, the authors concluded that SLGG3 glass can compete with commercial radiation shielding glasses.

End Image Defect Detection of Float Glass Based on Faster Region-Based Convolutional Neural Network

The float glass contains various defects for reasons of raw materials and production process. These defects can be observed on the end images of the glass. Since the defects are correlated with specific links of the production process, it is possible to discover the process problems by identifying the location and type of defects in end images. Based on faster region-based convolutional neural network (Faster RCNN), this paper proposes a deep learning method that improves the feature extraction network, and adds a Laplacian convolutional layer to preprocess the end images. Considering the defect features in end images, the anchor box size was adjusted to speed up the training. Besides, the lack of generalizability induced by small dataset was solved through data enhancement. With improved VGG16 as the feature extraction layer, a glass defect detection model was established, whose generalizability was improved through transfer learning. The experimental results show that the proposed model achieved a mean detection accuracy of 94% on actual test set, meeting the requirements for actual use in factories.

Approaches for Automated Monitoring and Evaluation of In Vitro Plant’s Morphometric Parameters

The article discusses approaches to solving the problem of precise and objective registration and assessment of In Vitro plant’s morphometric parameters. Evaluating the impact of different nutrient media composition at different stages of plant development can help to choose the right nutrient medium and get high-quality virus-free planting material. The process of manual measurements of plant growth parameters violates the microclimate formed for optimal plant development, assumes inaccuracies due to the influence of the human factor and requires processing large volumes of heterogeneous data. The described disadvantages of manual measurements justify the prospects of using modern information technologies, methods and tools for automation and modeling, which allows conducting monitoring, evaluating and predicting the growth and development of plants In Vitro. Today, there are various hardware and software systems for photographing plants and assessing their morphometric parameters. However, such systems do not take into account the location of plants behind the glass of the test tube, which introduces image distortions due to fogging and glass defects, non-linearity at the boundaries of the tube, glare and reflection of the glass. This article proposes a prototype of an automated system for photographing plants from different angles, creating a 3D volumetric model of the plant and calculating morphometric parameters at various stages of plant development.

Anomaly Detection Neural Network with Dual Auto-Encoders GAN and Its Industrial Inspection Applications.

Recently, researchers have been studying methods to introduce deep learning into automated optical inspection (AOI) systems to reduce labor costs. However, the integration of deep learning in the industry may encounter major challenges such as sample imbalance (defective products that only account for a small proportion). Therefore, in this study, an anomaly detection neural network, dual auto-encoder generative adversarial network (DAGAN), was developed to solve the problem of sample imbalance. With skip-connection and dual auto-encoder architecture, the proposed method exhibited excellent image reconstruction ability and training stability. Three datasets, namely public industrial detection training set, MVTec AD, with mobile phone screen glass and wood defect detection datasets, were used to verify the inspection ability of DAGAN. In addition, training with a limited amount of data was proposed to verify its detection ability. The results demonstrated that the areas under the curve (AUCs) of DAGAN were better than previous generative adversarial network-based anomaly detection models in 13 out of 17 categories in these datasets, especially in categories with high variability or noise. The maximum AUC improvement was 0.250 (toothbrush). Moreover, the proposed method exhibited better detection ability than the U-Net auto-encoder, which indicates the function of discriminator in this application. Furthermore, the proposed method had a high level of AUCs when using only a small amount of training data. DAGAN can significantly reduce the time and cost of collecting and labeling data when it is applied to industrial detection.

Synthesis and physical characterization of gamma irradiated cadmium phosphate glass modified by copper oxide

The irradiation of Cu-doped Cd-phosphate glass mediated by gamma rays shows the conversion of Cu2+ ↔ Cu+ ion with subsequent glass defects which leads to the loss of glass structure. The physical evaluations encompassing FTIR studies, X-ray diffraction analysis, UV spectrophotometer analysis, EPR, differential scanning colorimetry, and elastic properties are performed. The optical band gap (Eg) values for both direct and indirect transition and refractive index are evaluated for varying irradiation doses and correlated with the data obtained for optical spectra, ultrasonic wave velocities and elastic moduli. Further, the evaluations such as optical basicity and average electro negativity have been determined. The obtained optical basicity values increases with concurrent increase in the dose of radiation, however the values of electron negativity have not displayed such variations. After irradiation at various doses, there is a significant elevation in non-bridging oxygen (NBO) atoms which lead to decreased glass polymerization, reduced glass stability and aids the formation of copper ions aggregates. The above stated process highly alters the structure and optical properties of glass which is confirmed by the results of the optical absorption, FTIR and EPR spectral analysis. The force constant values and the bond strength between Cu–O are lower as compared to P–O, which further decreases the glass rigidity thus decrease in glass rigidity affects the ultrasonic velocity, Eg and Tg. The thermal stability values of studied glass decreased with subsequent increases in the dose of irradiation. Thus this glass displays specific optical characteristic and thus can be employed as a bandpass filters.

Visualization of defects in glass through pulsed thermography.

Pulsed thermography was exploited to identify the presence of glass defects in order to get an indication of the conservation status of archaeological glass. Indeed, the process of degradation in artifacts subjected to centuries of burial can be of great relevance. More specifically, we evaluated the potential of pulsed thermography to map the presence of flakes in archaeological glass. This was achieved by comparing different heating setups and signal-processing algorithms. Tests were carried out previously on glass mockups with surface defects and then on archaeological artifacts.