• The Journal of the Acoustical Society of Korea
• /
• v.39 no.4
• /
• pp.292-302
• /
• 2020

Machine learning algorithms have made immense contributions in various fields including sonar and radar applications. Recently developed Cycle-Consistency Generative Adversarial Network (CycleGAN), a variant of GAN has been successfully used for unpaired image-to-image translation. We present a modified CycleGAN for translation of underwater ship engine sounds with high perceptual quality. The proposed network is composed of an improved generator model trained to translate underwater audio from one vessel type to other, an improved discriminator to identify the data as real or fake and a modified cycle-consistency loss function. The quantitative and qualitative analysis of the proposed CycleGAN are performed on publicly available underwater dataset ShipsEar by evaluating and comparing Mel-cepstral distortion, pitch contour matching, nearest neighbor comparison and mean opinion score with existing algorithms. The analysis results of the proposed network demonstrate the effectiveness of the proposed network.

• Journal of the Korean Geosynthetics Society
• /
• v.21 no.2
• /
• pp.11-19
• /
• 2022

In this paper, we propose a GoogleNet transfer learning and CNN-LSTM combination method to improve the time-series prediction performance for crack detection using crack data captured inside the sewer pipes. LSTM can solve the long-term dependency problem of CNN, so spatial and temporal characteristics can be considered at the same time. The predictive performance of the proposed method is excellent in all test variables as a result of comparing the RMSE(Root Mean Square Error) for time series sections using the crack data inside the sewer pipe. In addition, as a result of examining the prediction performance at the time of data generation, the proposed method was verified that it is effective in predicting crack detection by comparing with the existing CNN-only model. If the proposed method and experimental results obtained through this study are utilized, it can be applied in various fields such as the environment and humanities where time series data occurs frequently as well as crack data of concrete structures.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.11a
• /
• pp.187-190
• /
• 2022

본 논문은 메타버스 등 인공지능 연계 증강/가상현실 부동 중계 플랫폼에서 부동산 영상 기반 매물 소개 시스템 구축에서 사생활 및 개인정보가 영상에 담기게 될 수 있는 위험이 존재하기에 부동산 영상 내의 개인정보 및 민감 정보를 인공지능 기술을 기반으로 검출하여 삭제해주고 복원해주는 인공지능 기술 연구개발을 목표로 하였다. 한국형 부동산 내 민감 object 를 정의하고, 최신 인공지능 딥러닝 기술 기반 민감 object detection 알고리즘을 연구 개발하며, 영상에서 삭제된 부분은 인공지능 기술을 기반으로 물체가 없는 실제 공간영상으로 복원해주는 영상복원 기술도 연구 개발하였다. 한국형 부동산 환경 (영상 촬영 조도, 디스플레이 스타일, 주변 가구 배치 등)에 맞는 인공지능 모델 구축을 위하여, 자체적으로 한국 영상 database 구축 및 Transfer learning for target domain adaptation 을 진행하였다. 제안된 알고리즘은 일반적인 환경에서 98%의 정확도와 challenge 환경에서 (occlusion 빛 반사, 저조도 등) 81%의 정확도를 보였다. 본 기술은 Proptech 분야에서 주목받고 있는 메타버스 기반 온라인 중계 서비스 기술을 활성화하기 위하여 기획되었으며, 특히 메타버스 부동산 중계 플랫폼의 활성화를 위하여 사생활 보호 측면에서 필요한 중요 기술을 인공지능 기술을 활용하여 연구 개발하였다.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.6
• /
• pp.1-12
• /
• 2024

Deep learning with faster and more accurate results requires large amounts of storage space and large computations. Accordingly, many studies are using hardware accelerators for quick and accurate calculations. However, the performance bottleneck is due to data movement between the hardware accelerators and the CPU. In this paper, we propose a data prefetch strategy that can efficiently reduce such operational bottlenecks. The core idea of the data prefetch strategy is to predict the data needed for the next task and upload it to local memory while the hardware accelerator (Matrix Multiplication Unit, MMU) performs a task. This strategy can be enhanced by using a dual buffer to perform read and write operations simultaneously. This reduces latency and execution time of data transfer. Through simulations, we demonstrate a 24% improvement in the performance of hardware accelerators by maximizing parallel processing with dual buffers and bottlenecks between memories with data prefetch.

• Journal of the Korean Society for Railway
• /
• v.19 no.3
• /
• pp.314-323
• /
• 2016

Overhead line voltage of DC railway traction substations has rising or falling characteristics depending on the acceleration and regenerative braking of the subway train loads. The suppression of this irregular fluctuation of the line voltage gives rise to improved energy efficiency of both the railway substation and the trains. This paper presents parameter estimation schemes using the RC circuit model for an overhead line voltage at a 1500V DC electric railway traction substation. A linear artificial neural network with a back-propagation learning algorithm was trained using the measurement data for an overhead line voltage and four feeder currents. The least square estimation method was configured to implement batch processing of these measurement data. These estimation results have been presented and performance analysis has been achieved through raw data simulation.

• Journal of the Korean association of regional geographers
• /
• v.10 no.4
• /
• pp.787-799
• /
• 2004

The industrial community in a metropolis has played an essential role to keep business in a large city. This paper aims to analyze the realities of trade network among enterprises at Buksungro Tools Commercial Cooperative in Daegu. The urban style industrial community is found at Buksungro in Daegu. There are more than one hundred small-sized enterprises. Major trades among enterprises are occurred within and/or by the area. The long-term trade networks within the Buksungro Tools Commercial Cooperative have played the key role to maintain the industrial community. The trade relationship has depended on business networks based on social capital rather than commercial mechanism. The trade networks have been established through credit transactions as well as handling troublesome orders. The trade networks help to transfer technology and to learn the tacit knowledge among firms. The long-term trade networks are more influenced by the social accessibility than spatial accessibility.

• The Journal of the Korea Contents Association
• /
• v.8 no.1
• /
• pp.116-127
• /
• 2008

In the $7^{th}$ education curriculum, computer education curriculum in the elementary and secondary schools is composited into the contents for the use of computers so that there are some limitations in teaching students the abilities for solving various problems of several areas using computers. Recently, the research has done to change the computer education curriculum for enhancing creativity and problem solving ability required by the future education. The contents of the main subject for enhancing them is of computer programming, however, there was not enough research on systematic programming education curriculum for leading to motivating learners and enhanced knowledge transfer to those learners. In this paper, we analysis the contents mathematics education curriculum with consecutive contents and in tight connection with computer education and then extract its programming related elements. Based on those, we propose a programming education curriculum with which we can teach systematically computer programing according to continual and systematic guidance in the elementary and secondary schools. And we develop a teaching model and learning guidance for teaching students programming methods with the computer programming education curriculum proposed in this paper.

• Journal of Intelligence and Information Systems
• /
• v.29 no.4
• /
• pp.257-269
• /
• 2023

There is currently a burgeoning demand for image synthesis from photos and videos using deep learning models. Existing video synthesis models solely extract motion information from the provided video to generate animation effects on photos. However, these synthesis models encounter challenges in achieving accurate lip synchronization with the audio and maintaining the image quality of the synthesized output. To tackle these issues, this paper introduces a novel framework based on an image animation approach. Within this framework, upon receiving a photo, a video, and audio input, it produces an output that not only retains the unique characteristics of the individuals in the photo but also synchronizes their movements with the provided video, achieving lip synchronization with the audio. Furthermore, a super-resolution model is employed to enhance the quality and resolution of the synthesized output.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.6 no.2
• /
• pp.149-158
• /
• 2005

In this paper, we have designed and implemented a portable e-tongue (electronic tongue) system using MACS (multi array chemical sensor) and PDA. The system embedded in PDA has merits such as comfortable user interface and data transfer by internet from on-site to remote computer. MACS was made up 7 electrodes (${NH_4}^+$, $Na^+$, $Cl^-$, ${NO_3}^-$, $K^+$, $Ca^{2+}$, $Na^+$, pH) and a reference electrode. For learning the system, we adapted the Levenberg-Marquardt algorithm based on the back-propagation, which could iteratively learned the pre-determined standard patterns, in e-tongue system. Conclusionally, the relationship between the standard patterns and unknown pattern can be easily analyzed. The e-tongue was applied to whiskeys and cognac (one high level whisky, one low level whiskey, two cognac) and 2 sample whiskeys for each standard patterns and unknown patterns. The relationship between the standard patterns and unknown patterns can be easily analyzed.

• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.127-146
• /
• 2022

Recently, as word embedding has shown excellent performance in various tasks of deep learning-based natural language processing, researches on the advancement and application of word, sentence, and document embedding are being actively conducted. Among them, cross-language transfer, which enables semantic exchange between different languages, is growing simultaneously with the development of embedding models. Academia's interests in vector alignment are growing with the expectation that it can be applied to various embedding-based analysis. In particular, vector alignment is expected to be applied to mapping between specialized domains and generalized domains. In other words, it is expected that it will be possible to map the vocabulary of specialized fields such as R&D, medicine, and law into the space of the pre-trained language model learned with huge volume of general-purpose documents, or provide a clue for mapping vocabulary between mutually different specialized fields. However, since linear-based vector alignment which has been mainly studied in academia basically assumes statistical linearity, it tends to simplify the vector space. This essentially assumes that different types of vector spaces are geometrically similar, which yields a limitation that it causes inevitable distortion in the alignment process. To overcome this limitation, we propose a deep learning-based vector alignment methodology that effectively learns the nonlinearity of data. The proposed methodology consists of sequential learning of a skip-connected autoencoder and a regression model to align the specialized word embedding expressed in each space to the general embedding space. Finally, through the inference of the two trained models, the specialized vocabulary can be aligned in the general space. To verify the performance of the proposed methodology, an experiment was performed on a total of 77,578 documents in the field of 'health care' among national R&D tasks performed from 2011 to 2020. As a result, it was confirmed that the proposed methodology showed superior performance in terms of cosine similarity compared to the existing linear vector alignment.