• ETRI Journal
• /
• 제30권3호
• /
• pp.394-402
• /
• 2008

Domain analysis is the process of analyzing related software systems in a domain to find their common and variable parts. In the case of device drivers, they are highly suitable for domain analysis because device drivers of the same domain are implemented similarly for each device and each system that they support. Considering this characteristic, this paper introduces a new approach to the domain analysis of device drivers. Our method uses a code clone detection technique to extract similarity among device drivers of the same domain. To examine the applicability of our method, we investigated whole device drivers of a Linux source. Results showed that many reusable similar codes can be discerned by the code clone detection method. We also investigated if our method is applicable to other kernel sources. However, the results show that the code clone detection method is not useful for the domain analysis of all kernel sources. That is, the applicability of the code clone detection method to domain analysis is a peculiar feature of device drivers.

• 한국컴퓨터정보학회논문지
• /
• 제23권1호
• /
• pp.17-24
• /
• 2018

With improvement of speech recognition and natural language processing, dialog systems are recently adapted to various service domains. It became possible to get desirable services by conversation through the dialog system, but it is still necessary to improve separate modules, such as domain detection, intention detection, named entity recognition, and out-of-domain detection, in order to achieve stable service offer. When it misclassifies an in-domain sentence of conversation as out-of-domain, it will result in poor customer satisfaction and finally lost business. As there have been relatively small number of studies related to the out-of-domain detection, in this paper, we introduce a new method using a hierarchical Dirichlet process and demonstrate the effectiveness of it by experimental results on Korean dataset.

• Journal of Communications and Networks
• /
• 제9권4호
• /
• pp.482-489
• /
• 2007

In this paper, two frequency domain signal processing techniques for pulse shape modulation(PSM) ultra wideband(UWB) systems are presented. Firstly, orthogonal detection of UWB PSM Hermite pulses in frequency domain is addressed. It is important because time domain detection by correlation-based receivers is severely degraded by many sources of distortion. Pulse-shape, the information conveying signal characteristic, is deformed by AWGN and shape-destructive addition of multiple paths from the propagation channel. Additionally, because of the short nature of UWB pulses, timing mismatches and synchronism degrade the performance of PSM UWB communication systems. In this paper, frequency domain orthogonality of the Hermite pulses is exploited to propose an alternative detection method, which makes possible efficient detection of PSM in dense multipath channel environments. Secondly, a ranging method employing the Cepstrum algorithm is proposed. This method is partly processed in the frequency domain and can be implemented without additional hardware complexity in the terminal.

• 대한임베디드공학회논문지
• /
• 제19권2호
• /
• pp.73-81
• /
• 2024

Recently, mobile manipulators have been utilized in agriculture industry for weed removal and harvest automation. This paper proposes a domain adaptive fruit detection method for harvest automation, by utilizing OWL-ViT model which is an open-vocabulary object detection model. The vision-language model can detect objects based on text prompt, and therefore, it can be extended to detect objects of undefined categories. In the development of deep learning models for real-world problems, constructing a large-scale labeled dataset is a time-consuming task and heavily relies on human effort. To reduce the labor-intensive workload, we utilized a large-scale public dataset as a source domain data and employed a domain adaptation method. Adversarial learning was conducted between a domain discriminator and feature extractor to reduce the gap between the distribution of feature vectors from the source domain and our target domain data. We collected a target domain dataset in a real-like environment and conducted experiments to demonstrate the effectiveness of the proposed method. In experiments, the domain adaptation method improved the AP50 metric from 38.88% to 78.59% for detecting objects within the range of 2m, and we achieved 81.7% of manipulation success rate.

• Quantitative Bio-Science
• /
• 제37권2호
• /
• pp.81-89
• /
• 2018

It is commonplace that high false detection rates interfere with immediate vision-based fire monitoring system. To circumvent this challenge, we propose a fire detection algorithm that can accommodate color variations of RGB in temporal domain, aiming at reducing false detection rates. Despite interrupting images (e.g., background noise and sudden intervention), the proposed method is proved robust in capturing distinguishable features of fire in temporal domain. In numerical studies, we carried out extensive real data experiments related to fire detection using 24 video sequences, implicating that the propose algorithm is found outstanding as an effective decision rule for fire detection (e.g., false detection rate <10%).

• 한국멀티미디어학회논문지
• /
• 제19권2호
• /
• pp.180-188
• /
• 2016

In this paper, we propose a fire detection algorithm based on motion information and color information analysis. Conventional fire detection algorithms have as main problem the difficulty to detect fire due to external light, intensity, background image complexity, and little fire diffusion. So we propose a fire detection algorithm that accurate and fast. First, it analyzes the motion information in video data and then set the first candidate. Second, it determines this domain after analyzing the color and the domain. This algorithm assures a fast fire detection and a high accuracy compared with conventional fire detection algorithms. Our algorithm will be useful to real-time fire detection in real world.

• 한국HCI학회:학술대회논문집
• /
• 한국HCI학회 2009년도 학술대회
• /
• pp.140-145
• /
• 2009

본 논문에서는 압축 영역에서 동작하는 조명 환경 변화에 강인한 얼굴 검출 방법을 제안한다. 기존 이미지 처리를 이용한 얼굴 검출 방법들은 주로 픽셀 기반 영역에서 이루어져 왔다. 그러나 컴퓨팅 파워와 저장 공간이 제한적인 로봇 환경에는 픽셀 기반 처리가 적합하지 않다. 또한 조명 환경의 변화는 안정된 얼굴 검출을 위해 해결되어야 하는 문제로 인식되어 왔다. 이러한 문제점들을 해결하기 위하여 본 논문에서는 압축 영역에서의 조명 효과 보상과 색 온도 변환을 이용한 색상 정보 조절 과정을 사용한 얼굴 검출 방법을 제안한다. 제안된 방법은 색상 정보 조절을 통하여 다양한 조명 환경에서 기존 방법에 비해 강인한 얼굴 검출을 보여준다.

• 한국음향학회지
• /
• 제30권2호
• /
• pp.73-79
• /
• 2011

본 논문에서는 음성 신호를 녹음하는 과정에서 발생하는 충격 잡음의 위치를 검출하는 새로운 알고리즘을 제안하였다. 제안한 방법은 충격 잡음의 주파수 축 특성을 반영하여 기존의 방법에 비해 높은 검출 정확도를 가지면서 음성의 피치를 충격 잡음과 구분하지 못하는 문제를 해결하였다. 또한, 시간 축, 주파수 축 파라미터의 단점을 상호 보완하여 false-alarm 문제를 최소화하는 시간-주파수 축 충격 잡음 검출 시스템을 제안하였다. 실제 녹음된 충격 잡음을 이용한 실험 결과, 제안한 시간-주파수 축 충격 잡음 검출기는 99.33 %의 가장 높은 검출 정확도와 1.49 %의 가장 낮은 false-alarm 비율을 나타내었다.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2015년도 춘계학술대회
• /
• pp.323-325
• /
• 2015

본 논문에서는 휴대용 방사능 검출기를 디지털로 구현하기 위하여 아날로그 부 시간영역에서 사용되던 Pulse shaper의 기능을 소프트웨어만으로 커버 할 수 있는 알고리즘을 제안한다.

• Journal of Information Technology Applications and Management
• /
• 제28권5호
• /
• pp.41-50
• /
• 2021

As COVID-19 spread widely, and rapidly, the number of misinformation is also increasing, which WHO has referred to this phenomenon as "Infodemic". The purpose of this research is to develop detection and rectification of COVID-19 misinformation based on Open-domain QA system and text similarity. 9 testing conditions were used in this model. For open-domain QA system, 6 conditions were applied using three different types of dataset types, scientific, social media, and news, both datasets, and two different methods of choosing the answer, choosing the top answer generated from the QA system and voting from the top three answers generated from QA system. The other 3 conditions were the Closed-Domain QA system with different dataset types. The best results from the testing model were 76% using all datasets with voting from the top 3 answers outperforming by 16% from the closed-domain model.