• 한국멀티미디어학회논문지
• /
• 제13권8호
• /
• pp.1115-1127
• /
• 2010

TFT-LCD 생산 과정에서 발생하는 결함을 정확하게 분류하여 결함 유형에 따라 폐기, 사용가능 등의 의사결정을 적절하게 내리는 것은 수율 증가 및 생산성 향상에 필수적인 요소이다. 본 논문에서는 TFT-LCD 생산 라인에서 획득한 결함 영상에 대하여 명암도 분포(intensity distribution) 및 결함 영상의 형태 특징(shape feature)을 분석하여 효과적으로 필름 결함 유형을 분류하는 기법을 제시한다. 본 연구에서는 먼저 필름 결함 영상을 결함 영역과 결함이 아닌 배경 영역으로 이진화하고, 결함 영역에서 결함의 선형성(linearity), 명암도 분포를 고려한 형태 특징 등의 여러 가지 특징을 분석하여 기준 영상(referential image) 데이터베이스를 구축하였으며, 분류하고자 하는 결함 영상과 데이터베이스에 저장된 기준 영상과의 매칭 비용 함수(matching cost function)를 정의하여 적절히 매칭시킴으로써 결함의 유형을 결정하였다. 제시한 기법의 성능을 검증하기 위하여 실제 TFT-LCD 생산 라인에서 획득한 결함 영상들을 대상으로 분류 실험을 수행하였으며, 실험 결과 생산 라인에서 이용할 수 있을 정도의 상당한 수준의 분류 정확도를 달성하였음을 보여주었다.

• 반도체디스플레이기술학회지
• /
• 제20권1호
• /
• pp.99-103
• /
• 2021

Warpage defect detecting of scaffold is very important in biosensor production. Because warpaged scaffold cause problem in cell culture. Currently, there is no detection equipment to warpaged scaffold. In this paper, we produced detection model for shape warpage detection using deep learning based CNN. We confirmed the shape of the scaffold that is widely used in cell culture. We produced scaffold specimens, which are widely used in biosensor fabrications. Then, the scaffold specimens were photographed to collect image data necessary for model manufacturing. We produced the detecting model of scaffold warpage defect using Densenet among CNN models. We evaluated the accuracy of the defect detection model with mAP, which evaluates the detection accuracy of deep learning. As a result of model evaluating, it was confirmed that the defect detection accuracy of the scaffold was more than 95%.

• 산업경영시스템학회지
• /
• 제46권2호
• /
• pp.1-12
• /
• 2023

To make semiconductor chips, a number of complex semiconductor manufacturing processes are required. Semiconductor chips that have undergone complex processes are subjected to EDS(Electrical Die Sorting) tests to check product quality, and a wafer bin map reflecting the information about the normal and defective chips is created. Defective chips found in the wafer bin map form various patterns, which are called defective patterns, and the defective patterns are a very important clue in determining the cause of defects in the process and design of semiconductors. Therefore, it is desired to automatically and quickly detect defective patterns in the field, and various methods have been proposed to detect defective patterns. Existing methods have considered simple, complex, and new defect patterns, but they had the disadvantage of being unable to provide field engineers the evidence of classification results through deep learning. It is necessary to supplement this and provide detailed information on the size, location, and patterns of the defects. In this paper, we propose an anomaly detection framework that can be explained through FCDD(Fully Convolutional Data Description) trained only with normal data to provide field engineers with details such as detection results of abnormal defect patterns, defect size, and location of defect patterns on wafer bin map. The results are analyzed using open dataset, providing prominent results of the proposed anomaly detection framework.

• 한국주조공학회지
• /
• 제37권6호
• /
• pp.181-185
• /
• 2017

In the present study, the effect of cooling rate on the formation of the porosity in the thick aluminum sand casting was investigated. Nowadays, due to considerations of weight and cost reduction, large scale thick aluminum casting has replaces steel frames for vacuum chambers for semiconductor production. Several thick aluminum castings were manufactured using chill with temperature measurements. The castings were inspected using 3D computed tomography in order to quantify the porosity defect density in the castings. Effects of the thickness of the chill on the porosity defect density were discussed.

• 한국생산제조학회지
• /
• 제6권4호
• /
• pp.46-53
• /
• 1997

This paper presents a method of detection localized defects on tapered roller bearing in main spindle of machine tool system. The feature vectors, i.e. statistical parameters, in time-domain analysis technique have been calculated to extract useful features from acoustic emission signals. These feature vectors are used as the input feature of an neural network to classify and detect bearing defects. As a results, the detection of bearing defect conditions could be sucessfully performed by using an neural network with statistical parameters of acoustic emission signals.

• 품질경영학회지
• /
• 제30권3호
• /
• pp.150-167
• /
• 2002

Normative theory suggests that control systems should be designed to complement management's objectives and strategies. However, few empirical studies have focused on how control systems have been modified to complement new manufacturing techniques, such as zero-defect strategy. The major purpose of this study is to examine the effects of the attributes of quality strategy and control system on performance. The previous studies have subdivided quality strategy into the ECL(economic conformance level) and ZD(zero-defect) strategy. To accomplish the purpose, this study empirically analysed the data based on the questionnaires from manufacturing department personnel of 67 Korean companies. The major results are as follows. The companies that had implemented the ZD strategy and its control system outperformed the companies that had implemented ECL strategy. Also, the ZD strategy differs from ECL strategy in control system, the criteria of performance evaluation and feedback frequence variable. There is no difference in quality performance monitoring and communication variable. From this results, although many companies have strived for continuous improvement of quality, it was restricted in improvement activities that dont's required much investment cost, quality performance monitoring and communication.

• Tribology and Lubricants
• /
• 제23권4호
• /
• pp.143-148
• /
• 2007

The surface defects of an aluminum tube for an OPC drum have been analyzed using a scanning microscopy(SEM) and an energy dispersive X-ray analyze.(EDX). The SEM/EDX system, which may provide good information on the surface defects and their distributions, provides an optical diameter of an impurity and a chemical composition. These are strongly related on the coated film thickness and quality of an OPC drum, which is a key element of a toner cartridge for a laser printer. The experimental results show that the local deformations, scratch wear, and flaws are produce the non-uniform coating layers, which may be removed by a manufacturing process of an aluminum tube. The major parameters on the coating quality of an OPC drum are the impurities of an aluminum tube such as silicon, oxygen, calcium, carbon, sulphur, chlorine, and others. These impurities may be removed by an ingot molding, extrusion and drawing, quality control, and packing processes with a strict manufacturing technology.

• 한국기계가공학회지
• /
• 제11권1호
• /
• pp.7-12
• /
• 2012

Chemical mechanical polishing(CMP) technology is faced with the challenge of processing new electronic materials. This paper focuses on the balance between chemical and mechanical reactions in the CMP process that is required to cope with a variety of electronic materials. The material properties were classified into the following categories: easy to abrade(ETA), difficult to abrade(DTA), easy to react(ETR) and difficult to react(DTR). The chemical and mechanical balance for the representative ETA-ETR, DTA-ETR, ETA-DTR and DTA-DTR materials was considered for defect-free surfaces. This paper suggests the suitable polishing methods and examples for each electronic material.

• 한국생산제조학회지
• /
• 제20권5호
• /
• pp.666-672
• /
• 2011

In this paper, we describes the development of automatic inspection system for detecting the defects on photovoltaic wafer by using machine vision. Until now, The defect inspection process was manually performed by operators. So these processes caused the produce of poorly-made articles and inaccuracy results. To improve the inspection accuracy, the inspection system is not only configured, but the image processing algorithm is also developed. The inspection system includes dimensional verification and pattern matching which compares a 2-D image of an object to a pattern image the method proves to be computationally efficient and accurate for real time application and we confirmed the applicability of the proposed method though the experience in a complex environment.

• 한국주조공학회지
• /
• 제24권2호
• /
• pp.63-70
• /
• 2004

Foundry techniquews for large steel casting depends on the skills of foundrymen considerably. Especially, the problem of reducing casring surface defects is difficult to clear numerically. Statistical analysis by using wuantification theory for hot tear and sand inclusion, and multiple regression analysis for dimensional defects have been shown to be examples of solving this difficulty. Many causes of surface defects can be evaluated by these analyses. These evaluations serve as the base data of defect reduction and contribute to the constant improvement of casting quality and quality enhancement activity. The system to perform quality enhancement activity was developed and it proved very useful for transfering foundry techniques and skills from the old to young generations.