• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.3
• /
• pp.59-66
• /
• 2016

This paper presents an empirical evaluation on dimensionality reduction strategies by which dissimilarity-based classifications (DBC) can be implemented efficiently. In DBC, classification is not based on feature measurements of individual objects (a set of attributes), but rather on a suitable dissimilarity measure among the individual objects (pair-wise object comparisons). One problem of DBC is the high dimensionality of the dissimilarity space when a lots of objects are treated. To address this issue, two kinds of solutions have been proposed in the literature: prototype selection (PS)-based methods and dimension reduction (DR)-based methods. In this paper, instead of utilizing the PS-based or DR-based methods, a way of performing DBC in Eigen spaces (ES) is considered and empirically compared. In ES-based DBC, classifications are performed as follows: first, a set of principal eigenvectors is extracted from the training data set using a principal component analysis; second, an Eigen space is expanded using a subset of the extracted and selected Eigen vectors; third, after measuring distances among the projected objects in the Eigen space using $l_p$-norms as the dissimilarity, classification is performed. The experimental results, which are obtained using the nearest neighbor rule with artificial and real-life benchmark data sets, demonstrate that when the dimensionality of the Eigen spaces has been selected appropriately, compared to the PS-based and DR-based methods, the performance of the ES-based DBC can be improved in terms of the classification accuracy.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.41 no.3
• /
• pp.229-235
• /
• 2021

Globally, due to climate change and urbanization, problems with water cycle destruction in urban areas frequently occur. In order to solve this problem, LID technique is being actively conducted the application in urban and research. In this study, some areas of the new city located in Busan was constructed using a widely used SWMM model to verify the effectiveness of the LID technique. This is to present a plan to maximize the efficiency of urban water cycle of the stormwater management target figure and the LID scale calculation method. In addition, the efficiency of runoff reduction using stormwater storage basin was analyzed in urban development projects. By calculating the scale of customized LID for each sub basin, the amount of runoff and peak runoff after LID application was reduced by 86.8 % and 69.5 %, respectively. Depending on the application of the stormwater storage basin, the reduction effect of peak runoff from 0.5 m³/s to 4.9 m³/s and delay effect of 8 minutes to 10 minutes was shown.

• Journal of Korea Society of Industrial Information Systems
• /
• v.27 no.2
• /
• pp.71-82
• /
• 2022

In order to replace the sampling tensile test of products produced in the tab welding process, which is one of the automotive battery manufacturing processes, vision inspectors are currently being developed and used. However, the vision inspection has the problem of inspection position error and the cost of improving it. In order to solve these problems, there are recent cases of applying deep learning technology. As one such case, this paper tries to examine the usefulness of applying Faster R-CNN, one of the deep learning technologies, to existing product inspection. The images acquired through the existing vision inspection machine are used as training data and trained using the Faster R-CNN ResNet101 V1 1024x1024 model. The results of the conventional vision test and Faster R-CNN test are compared and analyzed based on the test standards of 0% non-detection and 10% over-detection. The non-detection rate is 34.5% in the conventional vision test and 0% in the Faster R-CNN test. The over-detection rate is 100% in the conventional vision test and 6.9% in Faster R-CNN. From these results, it is confirmed that deep learning technology is very useful for detecting welding error of lead tabs in automobile batteries.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.6
• /
• pp.693-702
• /
• 2022

The purpose of this study is to present the problems of urban ecosystem imbalance in Seoul and the need to restore housing functions through comparison with Manhattan, New York, which grew earlier than Seoul and has been striving to solve urban problems and restore urban ecosystems. The residential status in urban commercial use areas was investigated through analysis of the total amount of residential space, characteristics of each type of residential building, and distribution of residential buildings. Through this, it can be seen that most of the space in Seoul's historic city center is biased toward commercial roads, and there are significantly insufficient high- and high-density mixed-use buildings suitable for urban dwellings compared to Manhattan's. In addition, it can be seen that the complex building in the historical city center of Seoul is located adjacent to the outskirts of the target site. In conclusion, Seoul's historical city center does not provide an appropriate residential space for creating an urban ecosystem, and it is necessary to come up with measures to restore housing functions.

• Journal of Broadcast Engineering
• /
• v.28 no.1
• /
• pp.31-41
• /
• 2023

Directly converting the focal depth and image size of computer-generated-hologram (CGH), which is obtained by calculating the interference pattern of light from the 3D image, is known to be quite difficult because of the less similarity between the CGH and the original image. This paper proposes a method for separately converting the each of focal length of the given CGH, which is composed of multi-depth images. Firstly, the proposed technique converts the 3D image reproduced from the CGH into a Light-Field (LF) image composed of a set of 2D images observed from various angles, and the positions of the moving objects for each observed views are checked using an object detection algorithm YOLOv5 (You-Only-Look-Once-version-5). After that, by adjusting the positions of objects, the depth-transformed LF image and CGH are generated. Numerical simulations and experimental results show that the proposed technique can change the focal length within a range of about 3 cm without significant loss of the image quality when applied to the image which have original depth of 10 cm, with a spatial light modulator which has a pixel size of 3.6 ㎛ and a resolution of 3840⨯2160.

• Journal of Platform Technology
• /
• v.11 no.2
• /
• pp.54-65
• /
• 2023

Due to the nature of electric vehicles, the batteries used for electric vehicles have a very large rated capacity. If an electric vehicle runs for a long time or an electric vehicle is abandoned due to a traffic accident, the electric vehicle battery becomes a waste battery. Even in vehicles that are being abandoned, the remaining capacity of waste batteries for electric vehicles is sufficient for other purposes. Waste batteries for automobiles are very expensive, so they need to be recycled and reused, but there was a problem that the standards for measuring the performance grade of waste batteries for recycling and reuse were insufficient. As a method for measuring the remaining capacity of waste battery, the most stable and reliable method is to measure the remaining capacity of battery using full charge and discharge. However, the inspection method by the full charging and discharging method varies depending on the capacity of the battery, but it takes more than a day to inspect, and many people are making great efforts to solve this problem. In this paper, an electric vehicle battery residual capacity analysis technique using voltage deviation between cells was studied and analyzed as a method to reduce inspection time for electric vehicle batteries. To this end, a full charging and discharging-based capacity measurement system was constructed, experimental data were collected using a nose or waste battery, and the correlation between the voltage deviation and the remaining capacity of the battery pack was analyzed to verify whether it can be used for battery inspection.

• Phonetics and Speech Sciences
• /
• v.15 no.3
• /
• pp.75-82
• /
• 2023

Automatic speech recognition (ASR) has been revolutionized with deep learning-based approaches, among which self-supervised learning methods have proven to be particularly effective. In this study, we aim to enhance the performance of OpenAI's Whisper model, a multilingual ASR system on the Korean language. Whisper was pretrained on a large corpus (around 680,000 hours) of web speech data and has demonstrated strong recognition performance for major languages. However, it faces challenges in recognizing languages such as Korean, which is not major language while training. We address this issue by fine-tuning the Whisper model with an additional dataset comprising about 1,000 hours of Korean speech. We also compare its performance against a Transformer model that was trained from scratch using the same dataset. Our results indicate that fine-tuning the Whisper model significantly improved its Korean speech recognition capabilities in terms of character error rate (CER). Specifically, the performance improved with increasing model size. However, the Whisper model's performance on English deteriorated post fine-tuning, emphasizing the need for further research to develop robust multilingual models. Our study demonstrates the potential of utilizing a fine-tuned Whisper model for Korean ASR applications. Future work will focus on multilingual recognition and optimization for real-time inference.

• Journal of Convergence for Information Technology
• /
• v.10 no.9
• /
• pp.1-7
• /
• 2020

In order to implement a smart factory for manufacturing innovation, more digital twins will be developed and applied gradually. In particular, simulation and optimization of digital twins makes it possible to support critical decision-making like a predictive maintenance of the equipment for manufacturing. In terms of a user perspective, this study suggests the conceptual framework of Digital Twin Management System (DTMS) for supporting the analytical and managerial activities for Digital Twins. We integrate the methods and structure of the area like Manufacturing Engineering, Decision Support Systems, and Optimization for developing the DTMS. The framework suggested in this study shows a typical DSS which consists of dialog management system, model management system and data management system. It also includes Analytical Digital Twins and simulations & optimization module. The framework is being applied in one of the most competitive and complex industrial sector. Also this study is meaningful to suggest a new direction of research.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.49 no.9
• /
• pp.251-258
• /
• 2012

Recently, NAND Flash memory structure is evolving from SDR (Single Data Rate) to high speed DDR(Double Data Rate) to fulfill the high performance requirement of SSD and SSS. Accordingly, the proper ways of transferring data that latches valid data stably and minimizing data skew between pins by using PHY(Physical layer) circuit techniques have became new issues. Also, rapid growth of speed in NAND flash increases the operating frequency and power consumption of NAND flash controller. Internal voltage variation margin of NAND flash controller will be narrowed through the smaller geometry and lower internal operating voltage below 1.5V. Therefore, the increase of power budge deviation limits the normal operation range of internal circuit. Affection of OCV(On Chip Variation) deteriorates the voltage variation problem and thus causes internal logic errors. In this case, it is too hard to debug, because it is not functional faults. In this paper, we propose new architecture that maintains the valid timing window in cost effective way under sudden power fluctuation cases. Simulation results show that the proposed technique minimizes the data skew by 379% with reduced area by 20% compared to using PHY circuits.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.6
• /
• pp.48-60
• /
• 2011

Nowadays, the 3D community is actively researching on 3D imaging and free-viewpoint video (FVV). The free-viewpoint rendering in multi-view video, virtually move through the scenes in order to create different viewpoints, has become a popular topic in 3D research that can lead to various applications. However, there are restrictions of cost-effectiveness and occupying large bandwidth in video transmission. An alternative to solve this problem is to generate virtual views using a single texture image and a corresponding depth image. A critical issue on generating virtual views is that the regions occluded by the foreground (FG) objects in the original views may become visible in the synthesized views. Filling this disocclusions (holes) in a visually plausible manner determines the quality of synthesis results. In this paper, a new approach for handling disocclusions using depth based inpainting algorithm in synthesized views is presented. Patch based non-parametric texture synthesis which shows excellent performance has two critical elements: determining where to fill first and determining what patch to be copied. In this work, a noise-robust filling priority using the structure tensor of Hessian matrix is proposed. Moreover, a patch matching algorithm excluding foreground region using depth map and considering epipolar line is proposed. Superiority of the proposed method over the existing methods is proved by comparing the experimental results.