• 한국컴퓨터정보학회논문지
• /
• 제28권3호
• /
• pp.35-43
• /
• 2023

운전 중 감정 인식은 사고를 예방하기 위해 꼭 필요한 과제이다. 더 나아가 자율 주행 시대에서 자동차는 모빌리티의 주체로 운전자와의 감정적인 소통이 더욱 요구되고 있으며 감정 인식 시장은 점점 확산되고 있다. 이에 따라 본 연구 방안에서는 수집하기 비교적 용이한 데이터인 심리데이터와 행동 데이터를 이용해 운전자의 감정을 분류하는 인공지능 모델을 개발하고자 한다. 오토인코더 모델을 통해 잠재 변수를 추출하고, 이를 본 분류 모델의 변수로 사용하였으며, 이는 성능 향상에 영향을 미침을 확인하였다. 또한 기존 뇌파 데이터를 포함했을 때 보다 본 논문이 제시하는 프레임워크를 사용하였을 때 성능이 향상됨도 확인하였다. 최종적으로 심리 및 개인정보데이터, 행동 데이터만을 통해 운전자의 감정 분류 정확도 81%와 F1-Score 80%를 달성하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권11호
• /
• pp.4084-4104
• /
• 2021

Visual place recognition (VPR) is a fundamental yet challenging task of mobile robot navigation and localization. The existing VPR methods are usually based on some pairwise similarity of image descriptors, so they are sensitive to visual appearance change and also computationally expensive. This paper proposes a simple yet effective four-step method that achieves adaptive convolutional neural network (CNN) landmark matching for VPR. First, based on the features extracted from existing CNN models, the regions with higher significance scores are selected as landmarks. Then, according to the coordinate positions of potential landmarks, landmark matching is improved by removing mismatched landmark pairs. Finally, considering the significance scores obtained in the first step, robust image retrieval is performed based on adaptive landmark matching, and it gives more weight to the landmark matching pairs with higher significance scores. To verify the efficiency and robustness of the proposed method, evaluations are conducted on standard benchmark datasets. The experimental results indicate that the proposed method reduces the feature representation space of place images by more than 75% with negligible loss in recognition precision. Also, it achieves a fast matching speed in similarity calculation, satisfying the real-time requirement.

• Structural Engineering and Mechanics
• /
• 제85권5호
• /
• pp.635-644
• /
• 2023

The design of honeycomb sandwich structures is often challenging because these structures can be tailored from a variety of possible cores and face sheets configurations, therefore, the design of sandwich structures is characterized as a time-consuming and complex task. A data-driven computational approach that integrates the analytical method and Artificial Neural Network (ANN) is developed by the authors to rapidly predict the design of sandwich structures for a targeted maximum structural deflection. The elaborated ANN reverse design approach is applied to obtain the thickness of the sandwich core, the thickness of the laminated face sheets, and safety factors for composite sandwich structure. The required data for building ANN model were obtained using the governing equations of sandwich components in conjunction with the Monte Carlo Method. Then, the functional relationship between the input and output features was created using the neural network Backpropagation (BP) algorithm. The input variables were the dimensions of the sandwich structure, the applied load, the core density, and the maximum deflection, which was the reverse input given by the designer. The outstanding performance of reverse ANN model revealed through a low value of mean square error (MSE) together with the coefficient of determination (R2) close to the unity. Furthermore, the output of the model was in good agreement with the analytical solution with a maximum error 4.7%. The combination of reverse concept and ANN may provide a potentially novel approach in designing of sandwich structures. The main added value of this study is the elaboration of a reverse ANN model, which provides a low computational technique as well as savestime in the design or redesign of sandwich structures compared to analytical and finite element approaches.

• 국제학술발표논문집
• /
• The 9th International Conference on Construction Engineering and Project Management
• /
• pp.1242-1242
• /
• 2022

Annually, several construction workers fall ill, are injured, or die due to heat-related exposure. The prevalence of work-related heat illness may rise and become an issue for workers operating in temperate climates, given the increase in frequency and intensity of heatwaves in the US. An increase in temperature negatively impacts physical exertion levels and mental state, thereby increasing the potential of accidents on the job site. To reduce the impact of heat stress on workers, it is critical to develop and implement measures for monitoring physical exertion levels and mental state in hot conditions. For this, limited studies have evaluated the utility of wearable biosensors in measuring physical exertion and mental workload in hot conditions. In addition, most studies focus solely on male participants, with little to no reference to female workers who may be exposed to greater heat stress risk. Therefore, this study aims to develop a process for objective and continuous assessment of worker physical exertion and mental workload using wearable biosensors. Physiological data were collected from eight (four male and four female) participants performing a simulated drilling task at 92oF and about 50% humidity level. After removing signal artifacts from the data using multiple filtering processes, the data was compared to a perceived muscle exertion scale and mental workload scale. Results indicate that biosensors' features can effectively detect the change in worker physical and mental state in hot conditions. Therefore, wearable biosensors provide a feasible and effective opportunity to continuously assess worker physical exertion and mental workload.

• 한국컴퓨터정보학회논문지
• /
• 제14권9호
• /
• pp.21-28
• /
• 2009

색상 극화란 주어진 텍스트 영역에서 글자색이 무엇인지를 결정하는 과정으로서 텍스트 추출을 위해서 선행되야 하는 작업이다. 본 논문에서는 텍스트 영역이 주어졌을 때 글자 영역을 추출하기 위한 색상 극화 기법을 제안한다. 제안된 방법은 글자 영역과 배경 영역에 대한 관찰을 바탕으로 두 영역 사이의 면적 비율과 표준편차비율의 관계를 색상 극화에 이용한다. 이를 위하여 그레이 스케일로 주어진 텍스트 영역을 Otsu의 방법으로 이진화하고 이진화된 두 영역을 각각 4-CC 레이블링한다. 레이블링된 두 그룹의 영역에 대해 각각 면적과 영역 중심으로부터의 거리에 대한 표준편차를 계산한 다음 두 그룹에서 면적이 가장 넓은 영역을 갖는 영역 사이의 면적 비와 표준편차가 가장 작은 영역들 사이의 표준편차 비를 이용하여 색상 극화를 수행한다. 다양한 폰트와 크기를 갖는 텍스트 영역에 대한 실험을 통해 제안된 방법이 강건하게 색상 극화를 수행함을 확인하였다.

• Genomics & Informatics
• /
• 제21권1호
• /
• pp.3.1-3.10
• /
• 2023

Characterization as well as prediction of the secondary and tertiary structure of hypothetical proteins from their amino acid sequences uploaded in databases by in silico approach are the critical issues in computational biology. Severe acute respiratory syndrome-associated coronavirus (SARS-CoV), which is responsible for pneumonia alike diseases, possesses a wide range of proteins of which many are still uncharacterized. The current study was conducted to reveal the physicochemical characteristics and structures of an uncharacterized protein Q6S8D9_SARS of SARS-CoV. Following the common flowchart of characterizing a hypothetical protein, several sophisticated computerized tools e.g., ExPASy Protparam, CD Search, SOPMA, PSIPRED, HHpred, etc. were employed to discover the functions and structures of Q6S8D9_SARS. After delineating the secondary and tertiary structures of the protein, some quality evaluating tools e.g., PROCHECK, ProSA-web etc. were performed to assess the structures and later the active site was identified also by CASTp v.3.0. The protein contains more negatively charged residues than positively charged residues and a high aliphatic index value which make the protein more stable. The 2D and 3D structures modeled by several bioinformatics tools ensured that the proteins had domain in it which indicated it was functional protein having the ability to trouble host antiviral inflammatory cytokine and interferon production pathways. Moreover, active site was found in the protein where ligand could bind. The study was aimed to unveil the features and structures of an uncharacterized protein of SARS-CoV which can be a therapeutic target for development of vaccines against the virus. Further research are needed to accomplish the task.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권10호
• /
• pp.2627-2642
• /
• 2023

Parkinson's disease (PD) is a typical, chronic neurodegenerative disease involving the concentration of dopamine, which can disrupt motor activity and cause different degrees of gait disturbance relevant to PD severity in patients. As current clinical PD diagnosis is a complex, time-consuming, and challenging task that relays on physicians' subjective evaluation of visual observations, gait disturbance has been extensively explored to make automatic detection of PD diagnosis and severity rating and provides auxiliary information for physicians' decisions using gait data from various acquisition devices. Among them, wearable sensors have the advantage of flexibility since they do not limit the wearers' activity sphere in this application scenario. In this paper, an attention-based temporal network (ATN) is designed for the time series structure of gait data (vertical ground reaction force signals) from foot sensor systems, to learn the discriminative differences related to PD severity levels hidden in sequential data. The structure of the proposed method is illuminated by Transformer Network for its success in excavating temporal information, containing three modules: a preprocessing module to map intra-moment features, a feature extractor computing complicated gait characteristic of the whole signal sequence in the temporal dimension, and a classifier for the final decision-making about PD severity assessment. The experiment is conducted on the public dataset PDgait of VGRF signals to verify the proposed model's validity and show promising classification performance compared with several existing methods.

• International Journal of Computer Science & Network Security
• /
• 제23권11호
• /
• pp.59-66
• /
• 2023

Text mining (TM) is most widely used to find patterns from various text documents. Cyber-bullying is the term that is used to abuse a person online or offline platform. Nowadays cyber-bullying becomes more dangerous to people who are using social networking sites (SNS). Cyber-bullying is of many types such as text messaging, morphed images, morphed videos, etc. It is a very difficult task to prevent this type of abuse of the person in online SNS. Finding accurate text mining patterns gives better results in detecting cyber-bullying on any platform. Cyber-bullying is developed with the online SNS to send defamatory statements or orally bully other persons or by using the online platform to abuse in front of SNS users. Deep Learning (DL) is one of the significant domains which are used to extract and learn the quality features dynamically from the low-level text inclusions. In this scenario, Convolutional neural networks (CNN) are used for training the text data, images, and videos. CNN is a very powerful approach to training on these types of data and achieved better text classification. In this paper, an Ensemble model is introduced with the integration of Term Frequency (TF)-Inverse document frequency (IDF) and Deep Neural Network (DNN) with advanced feature-extracting techniques to classify the bullying text, images, and videos. The proposed approach also focused on reducing the training time and memory usage which helps the classification improvement.

• 정보처리학회 논문지
• /
• 제13권4호
• /
• pp.166-173
• /
• 2024

본 연구에서는 얼굴 동영상에서 입술의 움직임과 음성 간의 동기화 탐지 방법을 제안한다. 기존의 연구에서는 얼굴 탐지 기술로 얼굴 영역의 바운딩 박스를 도출하고, 박스의 하단 절반 영역을 시각 인코더의 입력으로 사용하여 입술-음성 동기화 탐지에 필요한 시각적인 특징을 추출하였다. 본 연구에서는 입술-음성 동기화 탐지 모델이 음성 정보의 발화 영역인 입술에 더 집중할 수 있도록 사전 학습된 시각적 Attention 기반의 인코더 도입을 제안한다. 이를 위해 음성 정보 없이 시각적 정보만으로 발화하는 말을 예측하는 독순술(Lip-Reading)에서 사용된 Visual Transformer Pooling(VTP) 모듈을 인코더로 채택했다. 그리고, 제안 방법이 학습 파라미터 수가 적음에도 불구하고 LRS2 데이터 세트에서 다섯 프레임 기준으로 94.5% 정확도를 보임으로써 최근 모델인 VocaList를 능가하는 것을 실험적으로 증명하였다. 또, 제안 방법은 학습에 사용되지 않은 Acappella 데이터셋에서도 VocaList 모델보다 8% 가량의 성능 향상이 있음을 확인하였다.

• 한국음향학회지
• /
• 제29권5호
• /
• pp.332-338
• /
• 2010

그동안 공감이 대인관계에서 하는 역할은 꾸준히 밝혀져 왔다. 하지만 공감이 주관적 안녕감에 어떤 영향을 미치는지 밝히는 연구는 거의 없었다. 이에 따라 본 연구에서는 공감이 자기존중감을 매개로 주관적 안녕감에 영향을 미친다는 모형을 가정하고 검정하고자 하였다. 또한 허재홍, 이찬종 (2010)은 공감지수 (Empathy Quotient: EQ) 척도의 심리측정 속성은 밝혔으나 확인요인분석은 하지 않아 본 연구에서 확인요인분석을 하였다. 대학생 421명 (남학생 192명, 여학생 225명)을 대상으로 설문을 실시하였고, 구조방정식 모형과 경로분석을 하였다. 연구결과 EQ는 기존 연구와 마찬가지로 세 요인 (인지공감, 정서공감, 사회기술공감)으로 보는 것이 타당하였다. 또한 공감은 자기존중감을 매개로 주관적 안녕감에 영향을 미치고 있었다. 이 연구 결과에서 공감이 우리나라 문화에서 개인이 문화의 가치를 충족시킴으로써 자기존중감을 높이고 이는 다시 주관적 안녕감을 향상시킨다는 것을 알 수 있었다. 이와 함께 본 연구에서는 공감에 우리나라 문화 특성도 반영해야 한다는 사실도 시사되었다.