• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2008.10a
• /
• pp.410-414
• /
• 2008

음악 연구에 따른 컴퓨터의 역할이 점차 중요한 비중을 차지함에 따라 보다 효과적인 악보 인식 방법이 요구된다. 기존의 악보 인식 방법에서는 특정 수정 프로그램에서 만든 악보만 그 프로그램에서 재수정과 재생이 가능하다는 단점이 있다. 본 논문에서는 이러한 단점을 보완하기 위하여 이미 작성 되어있는 악보들을 자동으로 인식하고 재생을 할 수 있는 방법을 제안한다. 제안된 악보 인식 방법은 수평 히스토그램을 이용하여 악보 이미지의 오선을 제거한 후, Grassfire 알고리즘을 적용하여 잡음을 제거하고 악보 구성 기호들을 추출한다. 추출된 악보 구성 기호들은 악보 구성 기호의 특징을 이용하여 음표와 쉼표, 그 외의 기호들로 분리한다. 분리된 음표 기호들은 박자마다 다른 음표 형태의 특징을 이용하여 다시 세밀하게 분리하고 쉼표와 그 외의 기호들은 퍼지 ART 알고리즘을 적용하여 인식한다. 인식된 악보 구성 기호들을 이용하여 각각 정보를 저장하고 향후에 악보 구성 기호에 해당하는 음의 재생을 용이하게 한다. 제안된 악보 인식 방법의 성능을 평가하기 위해 50장의 악보 영상을 대상으로 실험한 결과, 본 논문에서 제시한 악보 영상의 인식 방법이 실험을 통해서 효율적인 것을 확인하였다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.05a
• /
• pp.82-85
• /
• 2014

퍼지 이진화는 영상에 대한 임계값을 원본 영상의 가장 밝은 픽셀과 가장 어두운 픽셀의 평균값으로 설정하고 이를 삼각형 타입의 소속 함수에 적용하여 영상을 이진화한다. 그러나 퍼지 이진화는 영상의 배경과 물체의 밝기 차이가 큰 경우에는 이진화가 효과적이지만 차이가 크지 않은 경우에는 소속 함수 구간을 효율적으로 설정할 수 없어 이진화를 효과적으로 할 수 없다. 따라서 본 논문에서는 이러한 문제점을 개선하기 위해 ART2 알고리즘을 적용하여 각 클러스터의 중심 값을 구한다. 그리고 각 클러스터의 중심 값에 해당하는 명암도를 이용하여 평균값을 구한 후, 이 평균값을 퍼지 이진화 방법에서 소속 함수 구간의 중간 값으로 설정하여 영상을 이진화 한다. 제안된 방법의 성능을 평가하기 위해 다양한 영상에서 제안된 방법과 기존의 퍼지 이진화 방법을 적용한 결과, 기존의 퍼지 이진화 방법보다 정보 손실이 적은 상태에서 영상이 이진화되는 것을 확인하였다.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.479-482
• /
• 2003

We propose a real time gas monitoring system for classifying various gases with different concentrations. Using thermal modulation of operating temperature of two sensors, we extract patterns of gases from the voltage across the load resistance. We adopt the relative resistance as a pre-processing method and an ART2 neural network as a pattern recognition method. The proposed method has been implemented in a real time embedded system with tin oxide gas sensors, TGS 2611, 2602 and an MSP430 ultra-low power microcontroller in the test chamber.

• Proceedings of the KIEE Conference
• /
• 2002.11c
• /
• pp.206-209
• /
• 2002

This paper proposes a fractal dimension method for measurement of degree of similarity between prototype pattern and input pattern at ART1 (Adaptive Resonance Theory 1) neural network. In order to confirm the validity of proposed method, comparison of the performance has made between the conventional method and the proposed method through simulation. The results show that the proposed method has considerably improved the performance.

• Journal of Information Processing Systems
• /
• v.17 no.1
• /
• pp.1-13
• /
• 2021

In this research, we study the problem of font image skeletonization using an end-to-end deep adversarial network, in contrast with the state-of-the-art methods that use mathematical algorithms. Several studies have been concerned with skeletonization, but a few have utilized deep learning. Further, no study has considered generative models based on deep neural networks for font character skeletonization, which are more delicate than natural objects. In this work, we take a step closer to producing realistic synthesized skeletons of font characters. We consider using an end-to-end deep adversarial network, SkelGAN, for font-image skeletonization, in contrast with the state-of-the-art methods that use mathematical algorithms. The proposed skeleton generator is proved superior to all well-known mathematical skeletonization methods in terms of character structure, including delicate strokes, serifs, and even special styles. Experimental results also demonstrate the dominance of our method against the state-of-the-art supervised image-to-image translation method in font character skeletonization task.

• Earthquakes and Structures
• /
• v.25 no.5
• /
• pp.369-383
• /
• 2023

Seismic resilience (SR) plays a vital role in evaluating and improving performance losses along with saving repair costs of structures from potential earthquakes. To further explore the developments, hotspots, and trend directions of SR, a total of 901 articles are obtained from the Web of Science (WoS) database. CiteSpace software is used to conduct a bibliometric analysis, which indicates an upward trend of publications in SR and explores the relationship of countries, journals, cited references, and keywords based on visual maps and detailed tables. Then, based on the results of the bibliometric analysis, a state-of-the-art review is conducted to further explore the current challenges and trend directions of SR. The trend directions can be divided into five categories: (a) SR assessments of infrastructure structures, (b) multi-hazard quantifications of SR, (c) seismic resilient structures, (d) refining and calibrating analytical models, and (e) multi-criteria decision-making frameworks for sustainability and SR.

• Current Optics and Photonics
• /
• v.7 no.5
• /
• pp.511-517
• /
• 2023

Pipelines play an important role in urban water supply and drainage, oil and gas transmission, etc. This paper presents a technique for pattern recognition of fiber optic vibration signals collected by a distributed vibration sensing (DVS) system using a deep learning residual network (ResNet). The optical fiber is laid on the pipeline, and the signal is collected by the DVS system and converted into a 64 × 64 single-channel grayscale image. The grayscale image is input into the ResNet to extract features, and finally the K-nearest-neighbors (KNN) algorithm is used to achieve the classification and recognition of pipeline damage.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.6
• /
• pp.1269-1274
• /
• 2014

In computer vision research, binarization procedure is one of the most frequently used tools to discriminate target objects from background in grey level binary image. Fuzzy binarization is a reliable technique in environment with high uncertainty such as medical image analysis by setting the threshold as the average of minimum and maximum brightness with triangle type fuzzy membership function. However, this technique is also known as contrast sensitive method thus its discrimination power is not so great when the image has low contrast difference between objects and backgrounds and suffer from information loss as a result. Thus, in this paper, we propose a fuzzy binarization using ART2 algorithm to handle such low contrast image analysis. Proposed ART2 algorithm is applied to determine the medium point of membership function in the fuzzy binarization paradigm. The proposed methods shows low information loss rate in our experiment.

• Safety and Health at Work
• /
• v.15 no.1
• /
• pp.33-41
• /
• 2024

Background: Dust generated during the processing of building materials enterprises can pose a serious health risk. The study aimed to compare and analyze the results of ART and the Monte Carlo model for the dust exposure assessment in building materials enterprises, to derive the application scope of the two models. Methods: First, ART and the Monte Carlo model were used to assess the exposure to dust in each of the 15 building materials enterprises. Then, a comparative analysis of the exposure assessment results was conducted. Finally, the model factors were analyzed using correlation analysis and the scope of application of the models was determined. Results: The results show that ART is mainly influenced by four factors, namely, localized controls, segregation, dispersion, surface contamination, and fugitive emissions, and applies to scenarios where the workplace information of the building materials enterprises is specific and the average dust concentration is greater than or equal to 1.5 mg/m³. The Monte Carlo model is mainly influenced by the dust concentration in the workplace of building materials enterprises and is suitable for scenarios where the dust concentration in the workplace of the building materials enterprises is relatively uniform and the average dust concentration is less than or equal to 6mg/m³. Conclusion: ART is most accurate when workplace information is specific and average dust concentration is > 1.5 mg/m³; whereas, The Monte Carlo model is the best when dust concentration is homogeneous and average dust concentration is < 6 mg/m³.

• Journal of Korea Multimedia Society
• /
• v.18 no.8
• /
• pp.980-987
• /
• 2015

Most of cultural area which people enjoy and consume is kitsch culture, though the culture is barely acknowledged. In this multimedia era, people create, publish and enjoy contents of 'webtoon(web+cartoon)', which is relatively faster and more convenient to enjoy comparing to published cartoon. Contrarily to its physical growth, the webtoon shows difficulty in advancing with other genres, is full of irritative factors and contents, and has tough time to become more sophisticated one. This study derived characteristics of kitsch in webtoon, suggested the way of webtoon as an art by analyzing conditions and usage motivation of webtoon. The conditions of kitsch are uniformity of mass consumption, popular vein, catharsis, vicarious satisfaction, immediate feedback-requiring image flood, and reproduction of techniques. The usage motivations of webtoon are studied as pursuit of information, entertainment, relaxation, socialization, convenience, and effectiveness. Usage motivation factors in webtoon and kitsch culture are almost identical. Contrary to its past of being underestimated and vulgarly considered, kitsch is acknowledged as an 'kitsch art'. By studying the process of becoming an art, the study has its purpose to suggest the experimental and developing way to make webtoon be acknowledged as 'webtoon art'.