• Journal of the Semiconductor & Display Technology
• /
• v.20 no.1
• /
• pp.64-69
• /
• 2021

In this paper, a method for improving the defect classification performance in low contrast, ununiformity and featureless steel plate surfaces has been studied based on deep convolution neural network and transfer-learning neural network. The steel plate surface images have low contrast, ununiformity, and featureless, so that the contrast between defect and defect-free regions are not discriminated. These characteristics make it difficult to extract the feature of the surface defect image. A classifier based on a deep convolution neural network is constructed to extract features automatically for effective classification of images with these characteristics. As results of the experiment, AlexNet-based transfer-learning classifier showed excellent classification performance of 99.43% with less than 160 seconds of training time. The proposed classification system showed excellent classification performance for low contrast, ununiformity, and featureless surface images.

• Journal of the Semiconductor & Display Technology
• /
• v.19 no.3
• /
• pp.1-6
• /
• 2020

In this paper, a method for improving the defect classification performance in steel plate surface has been studied, based on DWT(discrete wavelet transform) and SVM(support vector machine). Surface images of the steel plate have low contrast, uneven, and featureless, so that the contrast between defect and defect-free regions is not discriminated. These characteristics make it difficult to extract the feature of the surface defect image. In order to improve the characteristics of these images, a synthetic images based on discrete wavelet transform are modeled. Using the synthetic images, edge-based features are extracted and also geometrical features are computed. SVM was configured in order to classify defect images using extracted features. As results of the experiment, the support vector machine based classifier showed good classification performance of 94.3%. The proposed classifier is expected to contribute to the key element of inspection process in smart factory.

• Journal of Information Display
• /
• v.4 no.1
• /
• pp.1-8
• /
• 2003

Recent progress in both low pretilt and high pretilt defect free C1 surface stabilized ferroelectric liquid crystal (SSFLC) devices is reviewed. First, by numerical calculation to investigate the balance between surface azimuthal anchoring energy and bulk elastic energy within the confined chevron layer geometry of C1 and C2, it is possible to achieve a zigzag free C1 state by low azimuthal anchoring alignment with a low pretilt angle. The critical azimuthal anchoring coefficient for defect free C1 state is calculated. Its relationship with elastic constant, chevron angle as well as surface topography effect are also discussed. Second, using $5^{\circ}$ oblique SiO deposition alignment method a defect free, large memory angle, high contrast ratio and bistable C1 SSFLC display, which has potential for electronic paper applications has also been developed. The electrooptical properties and bistability of this device have been investigated. Various aspects of defect control are also discussed.

• Journal of Institute of Control, Robotics and Systems
• /
• v.14 no.5
• /
• pp.415-419
• /
• 2008

Detecting and correcting defects on LCD glasses early in the manufacturing process becomes important for panel makers to reduce the manufacturing costs and to improve productivity. Many attempts have been made and were successfully applied to detect and identify simple defects such as scratches, dents, and foreign objects on glasses. However, it is still difficult to robustly detect low-contrast defect region, called Mura or blemish area on glasses. Typically, these defect areas are roughly defined as relatively large, several millimeters of diameter, and relatively dark and/or bright region of low Signal-to-Noise Ratio (SNR) against background of low-frequency signal. The aim of this article is to present a robust algorithm to segment these blemish defects. Early 90's, a highly robust estimator, known as the Model-Fitting (MF) estimator was developed by X. Zhuang et. al. and have been successfully used in many computer vision application. Compared to the conventional Least-Square (LS) estimator the MF estimator can successfully estimate model parameters from a dataset of contaminated Gaussian mixture. Such a noise model is defined as a regular white Gaussian noise model with probability $1-\varepsilon$ plus an outlier process with probability $varepsilon$. In the sense of robust estimation, the blemish defect in images can be considered as being a group of outliers in the process of estimating image background model parameters. The algorithm developed in this paper uses a modified MF estimator to robustly estimate the background model and as a by-product to segment the blemish defects, the outliers.

• Journal of the Semiconductor & Display Technology
• /
• v.16 no.3
• /
• pp.14-19
• /
• 2017

In this paper, a fused illumination mechanism for detecting surface defects in steel plates was designed by applying the discriminant function that can differentiate the contrast of defects and non-defects. There is low contrast, non-uniformity, and no feature characteristics in steel plate surfaces. The fused illumination mechanism is devised, based on those characteristics. Optimum parameters of the fused illumination mechanism are determined by applying the discriminant function after acquiring the defect image in steel plate surfaces. The performance of the proposed mechanism is verified by experminets.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.11
• /
• pp.2085-2092
• /
• 2016

One of the most important properties of the IC substrate is that it should be uniform over large areas. Among the various physical approaches of wafer defect characterization, special attention is to be payed to the infrared techniques of inspection. In particular, a high spatial resolution, near infrared absorption method has been adopted to directly observe defects in semi-insulating GaAs. This technique, which relies on the mapping of infrared transmission, is both rapid and non-destructive. This method demonstrates in a direct way that the infrared images of GaAs crystals arise from defect absorption process. A new interpretation is presented for the observed reversal of contrast in the infrared absorption of nonuniformly distributed deep centers, related to EL2, in semi-insulating GaAs. The low temperature photoquenching experiment has demonstrated in a direct way that the contrast inverse images of GaAs wafers arise from both absorption and scattering mechanisms rather than charge re-distribution or local variation of bandgap.

• Journal of KIISE:Software and Applications
• /
• v.34 no.5
• /
• pp.447-454
• /
• 2007

TFT-LCD(Thin Film transistor liquid crystal display) has become actively used front panel display technology with increasing market. Intrinsically there is region of non uniformity with low contrast that to human eye is perceived as defect. As the gray level difference between the defect and the background is hardly distinguishable, conventional thresholding and edge detection techniques cannot be applied to detect the defect. Between the patterned and un-patterned LCD defects, this paper deals with un-patterned LCD defects by using independent component analysis, adaptive thresholding and skewness. Our method showed strong results even on noised LCD images and worked successfully on the manufacturing line.

• Imaging Science in Dentistry
• /
• v.37 no.4
• /
• pp.211-215
• /
• 2007

Purpose: The purpose of this study was to evaluate the effect of the kV on fractal dimension of trabecular bone in digital radiographs. Materials and Methods: 16 bone cores were obtained from patients who had taken partial resection of tibia due to accidents. Each bone core along with an aluminum step wedge was radiographed with an occlusal film at 0.08 sec and with the constant film-focus distance (32 cm). All radiographs were acquired at 60, 75, and 90 kV. A rectangular ROI was drawn at medial part, distal part, and the bone defect area of each bone core image according to each kV. The directional fractal dimension was measured using Fourier Transform spectrum, and the anisotropy was obtained using directional fractal dimension. The values were compared by the repeated measures ANOVA. Results : The fractal dimensions increased along with kV increase (p < 0.05). The anisotropy measurements did not show statistically significant difference according to kV change. The fractal dimensions of the bone defect areas of the bone cores have low values contrast to the non-defect areas of the bone cores. The anisotropy measurements of the bone defect areas were lower than those of the non-defect areas of the bone cores, but not statistically significant. Conclusion: Fractal analysis can notice a difference of a change of voltage of x-ray tube and bone defect or not. And anisotropy of a trabecular bone is coherent even with change of the voltage of x-ray tube or defecting off a part of bone.

• Journal of Family Resource Management and Policy Review
• /
• v.8 no.3
• /
• pp.33-46
• /
• 2004

This study examined the effects of consumers' perceptions and expectations about refund on consumers' behavior to refund and its outcome. Major results of this study could be summarized as follows. First, consumers' perceptions and expectations about refund were higher in the case of defective products. However, consumers' behavior to refund were generally passive resulting in the low chance of refund. When there was no defects in product, consumers perceptions and expectations for refund were generally lower resulting in passive and the low chance of refund. Second, when there was defects in product, both consumers' perceptions and expectation tot refund influenced their active behaviors for refund, but not in the successful outcome. In contrast, when there was no defect, those two variables impacted on the behaviors for refund positively In addition, the more active those behaviors requesting for refunds were, the more successful outcome in refund request.

• Tuberculosis and Respiratory Diseases
• /
• v.76 no.3
• /
• pp.136-140
• /
• 2014

Pulmonary artery sarcoma (PAS) is a rare, poorly differentiated malignancy arising from the intimal layer of the pulmonary artery. Contrast-enhanced chest computed tomography (CT) is a good diagnostic modality that shows a low-attenuation filling defect of the pulmonary artery in PAS patients. An 18-year-old man was referred to our hospital for the evaluation and management of cavitary pulmonary lesions that did not respond to treatment. A contrast-enhanced CT of the chest was performed, which showed a filling defect within the right interlobar pulmonary artery. The patient underwent a curative right pneumonectomy after confirmation of PAS. Although lung parenchymal lesions of PAS are generally nonspecific, it can be presented as cavities indicate pulmonary infarcts. Clinicians must consider the possibility of PAS as well as pulmonary thromboembolism in patients with pulmonary infarcts. So, we report the case with PAS that was diagnosed during the evaluation of cavitary pulmonary lesions and reviewed the literatures.