• 한국정보시스템학회지:정보시스템연구
• /
• 제32권4호
• /
• pp.149-162
• /
• 2023

Purpose In the textile industry, fabric defects significantly impact product quality and consumer satisfaction. This research seeks to enhance defect detection by developing a transformer-based deep learning image segmentation model for learning high-dimensional image features, overcoming the limitations of traditional image classification methods. Design/methodology/approach This study utilizes the ZJU-Leaper dataset to develop a model for detecting defects in fabrics. The ZJU-Leaper dataset includes defects such as presses, stains, warps, and scratches across various fabric patterns. The dataset was built using the defect labeling and image files from ZJU-Leaper, and experiments were conducted with deep learning image segmentation models including Deeplabv3, SegformerB0, SegformerB1, and Dinov2. Findings The experimental results of this study indicate that the SegformerB1 model achieved the highest performance with an mIOU of 83.61% and a Pixel F1 Score of 81.84%. The SegformerB1 model excelled in sensitivity for detecting fabric defect areas compared to other models. Detailed analysis of its inferences showed accurate predictions of diverse defects, such as stains and fine scratches, within intricated fabric designs.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권2호
• /
• pp.687-701
• /
• 2020

In the production of industrial fabric, it needs automatic real-time system to detect defects on the fabric for assuring the defect-free products flow to the market. At present, many visual-based methods are designed for detecting the fabric defects, but they usually lead to high false alarm. Base on this reason, we propose a Sobel operator combined with patch statistics (SOPS) algorithm for defects detection. First, we describe the defect detection model. mean filter is applied to preprocess the acquired image. Then, Sobel operator (SO) is applied to deal with the defect image, and we can get a coarse binary image. Finally, the binary image can be divided into many patches. For a given patch, a threshold is used to decide whether the patch is defect-free or not. Finally, a new image will be reconstructed, and we did a loop for the reconstructed image to suppress defects noise. Experiments show that the proposed SOPS algorithm is effective.

• 한국산업정보학회논문지
• /
• 제19권6호
• /
• pp.43-51
• /
• 2014

TFT-LCD 영상은 다양한 특성의 결함을 포함하고 있다. 배경 영역과의 휘도 차이가 커서 육안으로 식별 가능한 결함부터 휘도 차이가 매우 적어서 육안 검출이 어려운 한도성 결함까지 포함한다. 본 논문에서는 휘도 차이를 이용하여 결함 영역에 포함될 확률이 높은 결함 화소부터 순차적으로 단계를 진행하면서 결함 후보 화소를 검출하고, 검출된 후보 화소를 블랍으로 구성하여 블랍의 크기와 주변 영역과의 휘도차이를 이용한 기법을 통해 최종적으로 결함 영역과 잡음을 구분하여 검출하는 알고리즘을 제안한다. 제안한 알고리즘의 타당성을 확인하기 위해 다양한 결함을 포함하는 영상에 대한 실험 결과를 살펴봄으로써 신뢰도 높은 결함 검출 결과를 입증하였다.

• 한국컴퓨터그래픽스학회논문지
• /
• 제29권3호
• /
• pp.79-84
• /
• 2023

이미지 생성 모델은 다양한 분야에 적용되어 데이터 부족 문제와 시간 및 비용 문제를 극복하고 있다. 그러나 규칙적 패턴 이미지에서의 이미지 생성과 해당 데이터의 결함 검출에는 한계를 가진다. 본 논문에서는 이미지 생성 모델의 패턴 이미지 생성의 가능성을 확인하고 OLED 패널의 결함 검출을 위한 데이터 증강에 적용하였다. OLED 결함 검출 모델을 학습하기 위해 필요한 데이터는 OLED 패널의 높은 비용 문제로 실제 데이터 세트를 확보하기 어렵다. 그렇기 때문에 해당 데이터 세트를 확보한다 하더라도 여러가지 결함 유형을 정의하고 분류하는 작업이 필요하다. 이를 위한 가상의 기반 데이터 세트를 획득할 OLED 패널 결함 데이터 획득 시스템을 소개하고, 이미지 생성 모델로 해당 데이터를 증강한다. 또, 확산모델에서의 패턴 이미지 생성의 어려움을 확인하여 가능성을 제안하고, 이미지 생성 모델 이용한 데이터 증강 및 결함 검출 데이터 증강의 제한 사항을 개선하였다.

• Journal of Information Processing Systems
• /
• 제14권6호
• /
• pp.1331-1346
• /
• 2018

Aiming at the problem that the gradient-based edge detection operators are sensitive to the noise, causing the pseudo edges, a triqubit-state measurement-based edge detection algorithm is presented in this paper. Combing the image local and global structure information, the triqubit superposition states are used to represent the pixel features, so as to locate the image edge. Our algorithm consists of three steps. Firstly, the improved partial differential method is used to smooth the defect image. Secondly, the triqubit-state is characterized by three elements of the pixel saliency, edge statistical characteristics and gray scale contrast to achieve the defect image from the gray space to the quantum space mapping. Thirdly, the edge image is outputted according to the quantum measurement, local gradient maximization and neighborhood chain code searching. Compared with other methods, the simulation experiments indicate that our algorithm has less pseudo edges and higher edge detection accuracy.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2022년도 춘계학술대회
• /
• pp.108-111
• /
• 2022

제품의 결함 탐지를 위한 머신 비전 시스템에 딥러닝을 적용하기 위해서는 다양한 결함 사례에 대한 방대한 학습 데이터가 필요하다. 하지만 실제 제조 산업에서는 결함의 종류에 따른 데이터 불균형이 생기기 때문에 결함 사례를 일반화할 수 있을 만큼의 제품 이미지를 수집하기 위해서는 많은 시간이 소요된다. 본 논문에서는 적은 데이터로도 학습이 가능한 샴 신경망을 제품 결함 탐지에 적용하고, 제품 결함 이미지 데이터의 속성을 고려하여 이미지 쌍 구성법과 대조 손실 함수를 수정하였다. AUC-ROC로 샴 신경망의 임베딩 성능을 간접적으로 확인한 결과, 같은 제품끼리만 쌍을 구성하고 결함이 있는 제품 간에는 쌍을 구성하였을 때, 그리고 지수 대조 손실로 학습하였을 때 좋은 임베딩 성능을 보였다.

• 제어로봇시스템학회논문지
• /
• 제17권10호
• /
• pp.1044-1050
• /
• 2011

The quality of brazing joints is one of the most important factors that have an effect on the performance of the brazing joint-based heat exchangers with the growing use in industry recently. Therefore, it is necessary to inspect the brazing joints in order to guarantee the performance of the heat exchangers. This paper presents a non-destructive method to inspect the brazing joints of the heat exchangers using X-ray. Firstly, X-ray cross-sectional images of the brazing joints are obtained by using CT (Computerized Tomography) technology. Cross-sectional image from CT is more useful to detect the inner defects than the traditional transmitted X-ray image. Secondly, the acquired images are processed by an algorithm proposed for the defect detection of brazing joint. Finally, two types of brazing joint are examined in a series of experiments to detect the defects in brazing joints. The experimental results show that the proposed algorithm is effective for defect detection of the brazing joints in heat exchangers.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1999년도 하계학술대회 논문집 B
• /
• pp.668-670
• /
• 1999

We have developed a automatic surface inspection system for cold Rolled strips in steel making process for several years. We have experienced the various kinds of surface inspection systems, including linear CCD camera type and the laser type inspection system which was installed in cold rolled strips production lines. But, we did not satisfied with these inspection systems owing to insufficient detection and classification rate, real time processing performance and limited line speed of real production lines. In order to increase detection and computing power, we have used the Dark Field illumination with Infra_Red LED, Bright Field illumination with Xenon Lamp, Parallel Computing Processor with Area typed CCD camera and full software based image processing technique for the ease up_grading and maintenance. In this paper, we introduced the automatic inspection system and real time image processing technique using the Object Detection, Defect Detection, Classification algorithms. As a result of experiment, under the situation of the high speed processed line(max 1000 meter per minute) defect detection is above 90% for all occurred defects in real line, defect name classification rate is about 80% for most frequently occurred 8 defect, and defect grade classification rate is 84% for name classified defect.

• Journal of the Korean Physical Society
• /
• 제73권11호
• /
• pp.1644-1649
• /
• 2018

Aiming at infrared thermal images with different buried depth defects, we study a variety of image segmentation algorithms based on the threshold to develop global search ability and the ability to find the defect area accurately. Firstly, the iterative thresholding method, the maximum entropy method, the minimum error method, the Ostu method and the minimum skewness method are applied to image segmentation of the same infrared thermal image. The study shows that the maximum entropy method and the minimum error method have strong global search capability and can simultaneously extract defects at different depths. However none of these five methods can accurately calculate the defect area at different depths. In order to solve this problem, we put forward a strategy of "divide and conquer". The infrared thermal image is divided into several local thermal maps, with each map containing only one defect, and the defect area is calculated after local image processing of the different buried defects one by one. The results show that, under the "divide and conquer" strategy, the iterative threshold method and the Ostu method have the advantage of high precision and can accurately extract the area of different defects at different depths, with an error of less than 5%.

• Nuclear Engineering and Technology
• /
• 제39권5호
• /
• pp.637-646
• /
• 2007

An UET (ultrasound excited thermography) has been used for several years for a remote non-destructive testing in the automotive and aircraft industry. It provides a thermo sonic image for a defect detection. A thermograhy is based On a propagation and a reflection of a thermal wave, which is launched from the surface into the inspected sample by an absorption of a modulated radiation. For an energy deposition to a sample, the UET uses an ultrasound excited vibration energy as an internal heat source. In this paper the applicability of the UET for a realtime defect detection is described. Measurements were performed on two kinds of pipes made from a copper and a CFRP material. In the interior of the CFRP pipe (70mm diameter), a groove (width - 6mm, depth - 2.7mm, and length - 70mm) was engraved by a milling. In the case of the copper pipe, a defect was made with a groove (width - 2mm, depth - 1mm, and length - 110 mm) by the same method. An ultrasonic vibration energy of a pulsed type is injected into the exterior side of the pipe. A hot spot, which is a small area around the defect was considerably heated up when compared to the other intact areas, was observed. A test On a damaged copper pipe produced a thermo sonic image, which was an excellent image contrast when compared to a CFRP pipe. Test on a CFRP pipe with a subsurface defect revealed a thermo sonic image at the groove position which was a relatively weak contrast.