• The Journal of Information Systems
• /
• v.32 no.4
• /
• 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.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.5
• /
• pp.236-242
• /
• 2024

Software Defect Prediction (SDP) enhances the efficiency of software development by proactively identifying modules likely to contain errors. A major challenge in SDP is improving prediction performance. Recent research has applied deep learning techniques to the field of SDP, with the SAINT model particularly gaining attention for its outstanding performance in analyzing structured data. This study compares the SAINT model with other leading models (XGBoost, Random Forest, CatBoost) and investigates the latest deep learning techniques applicable to SDP. SAINT consistently demonstrated superior performance, proving effective in improving defect prediction accuracy. These findings highlight the potential of the SAINT model to advance defect prediction methodologies in practical software development scenarios, and were achieved through a rigorous methodology including cross-validation, feature scaling, and comparative analysis.

• KIISE Transactions on Computing Practices
• /
• v.22 no.12
• /
• pp.681-686
• /
• 2016

Software defect prediction is an important factor in efficient project management and success. The severity of the defect usually determines the degree to which the project is affected. However, existing studies focus only on the presence or absence of a defect and not the severity of defect. In this study, we proposed an ensemble model using FCM based on defect severity. The severity of the defect of NASA data set's PC4 was reclassified. To select the input column that affected the severity of the defect, we extracted the important defect factor of the data set using Random Forest (RF). We evaluated the performance of the model by changing the parameters in the 10-fold cross-validation. The evaluation results were as follows. First, defect severities were reclassified from 58, 40, 80 to 30, 20, 128. Second, BRANCH_COUNT was an important input column for the degree of severity in terms of accuracy and node impurities. Third, smaller tree number led to more variables for good performance.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.7
• /
• pp.1429-1434
• /
• 2017

In automatic defect detection problem for spectacle lenses, it is important to extract lens area accurately. Many existing detection methods fail to do it due to insufficient minute noise removal. In this paper, we propose an automatic defect detection method using Bresenham algorithm and adaptive binarization strategy. After usual average binarization, we apply Bresenham algorithm that has the power in extracting ellipse shape from image. Then, adaptive binarization strategy is applied to the critical minute noise removal inside the lens area. After noise removal, We can also compute the influence factor of the defect based on the fuzzy logic with two membership functions such as the size of the defect and the distance of the defect from the center of the lens. In experiment, our method successfully extracts defects in 10 out of 12 example images that include CHEMI, MID, HL, HM type lenses.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.47 no.11
• /
• pp.23-29
• /
• 2010

This paper presents a method for constructing a memory system using defective memory chips comprising faulty storage blocks. The three-dimensional memory system introduced here employs a die-stacked structure of faulty memory chips. Signals lines passing through the through-silicon-vias (TSVs) connect chips in the defect tolerant structure. Defective chips are classified into several groups each group comprising defective chips having faulty blocks at the same location. A defect tolerant memory system is constructed using chips from different groups. Defect-free storage blocks from spare chips replace faulty blocks using additional routing circuitry. The number of spare chips for defect tolerance is $s={\ulcorner}(k{\times}n)/(m-k){\urcorner}$ to make a system defect tolerant for (n+s) chips with k faulty blocks among m independently addressable blocks.

• Journal of Institute of Control, Robotics and Systems
• /
• v.2 no.4
• /
• pp.337-343
• /
• 1996

In this paper, we investigate the defect detection by making use of pre-made reference image data and classify the defects by using the artificial neural network. The approach is composed of three main parts. The first step consists of a proper generation of two reference image data by using a low level morphological technique. The second step proceeds by performing three times logical bit operations between two ready-made reference images and just captured image to be tested. This results in defects image only. In the third step, by extracting four features from each detected defect, followed by assigning them into the input nodes of an already trained artificial neural network we can obtain a defect class corresponding to the features. All of the image data are formed in a bit level for the reduction of data size as well as time saving. Experimental results show that proposed algorithms are found to be effective for flexible defect detection, robust classification, and high speed process by adopting a simple logic operation.

• Journal of the Optical Society of Korea
• /
• v.19 no.6
• /
• pp.619-628
• /
• 2015

In this paper, we consider a method for estimating a stripe-type defect and the reconstruction of a defect-free L/S type mask used in lithography. Comparing diffraction patterns of defected and defect-free masks, we derive equations for the estimation of the location and size of the defect. We construct an analytical model for this problem and derive closed form equations to determine the location and size using phase retrieval problem solving techniques. Consequently, we develop an algorithm that determines a defect-free mask pattern. An example shows the validity of the equations.

• 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.

• Proceedings of the KIEE Conference
• /
• summer
• /
• pp.887-889
• /
• 2004

This paper presents an analysis of RPC probe signal in steam generator tube with defect using finite element method. Impedance signal is calculated according to the depth variation of defect in tube and change of frequency in same defect. As the depth of the defect and the operating frequency is increased, the magnitude of the signal is increased. From the result of this paper, we can obtain the information by the effect of defect and frequency.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2010.05a
• /
• pp.135-138
• /
• 2010

In the construction industry, construction management system using information technology has been applied diversely to increase productivity. Although IT device such as PDA, RFID, Barcode, wireless network and web camera has been introduced in construction site, the usage of these device is restricted, because these cause engineer to do additional work. The suggested process using Digital Pen in this study, which can lessen engineer's additional work, get defect data promptly. Also accumulated data is utilized effectively for analyzing construction site. The purpose of this study is to introduce Digital Pen System as a means of construction data acquisition for improving productivity.