• 디지털산업정보학회논문지
• /
• 제19권3호
• /
• pp.213-220
• /
• 2023

Face recognition is a very important process in video monitoring and is a type of biometric technology. It is mainly used for identification and security purposes, such as ID cards, licenses, and passports. The recognition process has many variables and is complex, so development has been slow. In this paper, we proposed a face recognition method using CNN, which has been re-examined due to the recent development of computers and algorithms, and compared with the feature comparison method, which is an existing face recognition algorithm, to verify performance. The proposed face search method is divided into a face region extraction step and a learning step. For learning, face images were standardized to 50×50 pixels, and learning was conducted while minimizing unnecessary nodes. In this paper, convolution and polling-based techniques, which are one of the deep learning technologies, were used for learning, and 1,000 face images were randomly selected from among 7,000 images of Caltech, and as a result of inspection, the final recognition rate was 98%.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2021년도 춘계학술대회
• /
• pp.38-40
• /
• 2021

의료 데이터 분야는 레코드 수는 많지만 응답값이 없기 때문에 인공지능을 적극적으로 활용하지 못하고 있다. 이러한 문제점을 해결하기 위해 자기지도학습(Self-Supervised learning)을 의료 분야에 적용하는 연구가 등장하고 있다. 자기지도학습은 model이 레이블링이 없는 데이터의 semantic 표현을 이해할 수 있도록 pretext task와 supervision을 학습한다. 그러나, 자기지도학습의 성능은 pretext task로 학습한 표현에 의존하므로 데이터의 특성에 적합한 pretext task를 정의할 필요가 있다. 따라서 본 논문에서는 의학 데이터 중 활용도가 높은 x-ray 이미지에 적용할 수 있는 pretext task를 실험적으로 탐색하고 그 결과를 분석한다.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2020년도 춘계학술발표대회
• /
• pp.222-223
• /
• 2020

최근 악성코드로 인한 피해가 증가하고 있다. 악성코드는 악성코드가 속한 종류에 따라서 대응하는 방법도 다르기 때문에 악성코드를 종류별로 분류하는 연구도 중요하다. 기존에는 악성코드 시각화 과정을 통해서 생성된 악성코드의 글로벌 이미지를 사용해 악성코드를 각 종류별로 분류한다. 글로벌 이미지를 악성코드로부터 추출한 바이너리 정보를 사용해서 생성한다. 하지만, 글로벌 이미지만을 사용해서 악성코드를 각 종류별로 분류하는 경우 악성코드의 종류별로 중요한 특징을 고려하기 않기 때문에 분류 정확도가 떨어진다. 본 논문에서는 악성코드의 글로벌 이미지에 악성코드의 종류별 특징을 나타내기 위한 로컬 특징 기반 글로벌 이미지를 사용한 악성코드 분류 방법을 제안한다. 첫 번째, 악성 코드로부터 바이너리를 추출하고 추출된 바이너리를 사용해서 글로벌 이미지를 생성한다. 두 번째, 악성 코드로부터 로컬 특징을 추출하고 악성코드의 종류별 핵심 로컬 특징을 단어-역문서 빈도(Term Frequency Inverse Document Frequency, TFIDF) 알고리즘을 사용해 선택한다. 세 번째, 생성된 글로벌 이미지에 악성코드의 패밀리별 핵심 특징을 픽셀화해서 적용한다. 네 번째, 생성된 로컬 특징 기반 글로벌 이미지를 사용해서 컨볼루션 모델을 학습하고, 학습된 컨볼루션 모델을 사용해서 악성코드를 각 종류별로 분류한다.

• 문화기술의 융합
• /
• 제9권2호
• /
• pp.245-250
• /
• 2023

이 논문은 국내 영화 <거울속으로>와 리메이크인 <미러>를 비교 분석하는 것이다. 두 영화는 거울에 반사된 이미지와 실제와의 차이를 부각시키는 21세기 초 국내 영화의 경향을 예시한다. 기본적인 설정과 플롯에는 유사성이 있지만, 내러티브를 이끌어가는 동기 등에서 주요한 차이점들이 있다. <거울속으로>는 원혼설화의 서사전통을 따르는데, 권선징악의 구도에 사회 비판 의식이 내포되어 있다. 거울을 통한 악령의 사악한 작용이 지배적인 <미러>는 서양의 공포영화의 전통을 따르고 있다. 이 리메이크는 동양문화를 근간으로 하는 서사를 기독교 문화가 지배적인 서구적 관점에서 재해석하고 재창조하는 한 방식을 예시해준다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권5호
• /
• pp.1671-1686
• /
• 2022

Face manipulation tools represented by Deepfake have threatened the security of people's biological identity information. Particularly, manipulation tools with deep learning technology have brought great challenges to Deepfake detection. There are many solutions for Deepfake detection based on traditional machine learning and advanced deep learning. However, those solutions of detectors almost have problems of poor performance when evaluated on different quality datasets. In this paper, for the sake of making high-quality Deepfake datasets, we provide a preprocessing method based on the image pixel matrix feature to eliminate similar images and the residual channel attention network (RCAN) to resize the scale of images. Significantly, we also describe a Deepfake detector named Cascaded-Hop which is based on the PixelHop++ system and the successive subspace learning (SSL) model. By feeding the preprocessed datasets, Cascaded-Hop achieves a good classification result on different manipulation types and multiple quality datasets. According to the experiment on FaceForensics++ and Celeb-DF, the AUC (area under curve) results of our proposed methods are comparable to the state-of-the-art models.

• 품질경영학회지
• /
• 제51권2호
• /
• pp.223-245
• /
• 2023

Purpose: This study aim to identify the trends in AI-based PHM technology that can enhance reliability and minimize costs. Furthermore, this research provides valuable guidelines for future studies in various industries Methods: In this study, I collected and selected AI-based PHM studies, established classification criteria, and analyzed research trends based on classified fields and techniques. Results: Analysis of 125 domestic studies revealed a greater emphasis on machinery in both diagnosis and prognosis, with more papers dedicated to diagnosis. various algorithms were employed, including CNN for image diagnosis and frequency analysis for signal data. LSTM was commonly used in prognosis for predicting failures and remaining life. Different industries, data types, and objectives required diverse AI techniques, with GAN used for data augmentation and GA for feature extraction. Conclusion: As studies on AI-based PHM continue to grow, selecting appropriate algorithms for data types and analysis purposes is essential. Thus, analyzing research trends in AI-based PHM is crucial for its rapid development.

• 한국건축시공학회:학술대회논문집
• /
• 한국건축시공학회 2022년도 가을 학술논문 발표대회
• /
• pp.47-48
• /
• 2022

The positioning technology that measures the position of a person or object is a key technology to deal with the location of the real coordinate system or converge the real and virtual worlds, such as digital twins, augmented reality, virtual reality, and autonomous driving. In estimating the location of a person or object at an indoor construction site, there are restrictions that it is impossible to receive location information from the outside, the communication infrastructure is insufficient, and it is difficult to install additional devices. Therefore, this study tested the direct sparse odometry algorithm, one of the visual Simultaneous Localization and Mapping (vSLAM) that estimate the current location and surrounding map using only image information, at an indoor construction site and analyzed its applicability as an indoor positioning technology. As a result, it was found that it is possible to properly estimate the surrounding map and the current location even in the indoor construction site, which has relatively few feature points. The results of this study can be used as reference data for researchers related to indoor positioning technology for construction sites in the future.

• Communications for Statistical Applications and Methods
• /
• 제30권6호
• /
• pp.589-603
• /
• 2023

Deep generative models target to infer the underlying true data distribution, and it leads to a huge success in generating fake-but-realistic data. Regarding such a perspective, the data attributes can be a crucial factor in the data generation process since non-existent counterfactual samples can be generated by altering certain factors. For example, we can generate new portrait images by flipping the gender attribute or altering the hair color attributes. This paper proposes counterfactual disentangled variational autoencoder generative adversarial networks (CDVAE-GAN), specialized for data attribute level counterfactual data generation. The structure of the proposed CDVAE-GAN consists of variational autoencoders and generative adversarial networks. Specifically, we adopt a Gaussian variational autoencoder to extract low-dimensional disentangled data features and auxiliary Bernoulli latent variables to model the data attributes separately. Also, we utilize a generative adversarial network to generate data with high fidelity. By enjoying the benefits of the variational autoencoder with the additional Bernoulli latent variables and the generative adversarial network, the proposed CDVAE-GAN can control the data attributes, and it enables producing counterfactual data. Our experimental result on the CelebA dataset qualitatively shows that the generated samples from CDVAE-GAN are realistic. Also, the quantitative results support that the proposed model can produce data that can deceive other machine learning classifiers with the altered data attributes.

• 한국환경과학회지
• /
• 제32권12호
• /
• pp.979-985
• /
• 2023

After the outbreak of COVID-19, the demand for home-enjoyable video games has surged, driven by extended pandemic conditions and resulting in rapid graphic technology advancements. Consequently, games with diverse expression techniques have emerged, captivating players. Virtual Reality (VR) environments predominantly feature natural landscapes, with advancing graphic technology enabling lifelike scenes. The rise in individuals seeking solace through natural elements in games has followed suit. As VR technology and metaverse interest grow, more people are exposed to digital imagery. However, evidence on the influence of graphic expression methods on emotional response to that is lacking. Our study examined user emotional responses, focusing on natural landscapes in digital graphics of games. Analyzing a group of 47 young adults as frequent digital media consumers, we studied reactions to different image styles (Realism, Semi-Realism, Stylized). In the analysis, Realism-style images were perceived the most positively, while emotional responses to natural landscapes with different graphical expressions showed no significant differences. Results suggest that recognizing digital natural landscapes may outweigh expression style impacting the evaluation of digital nature. This study's empirical analysis enhances the understanding of digital nature's application to actual situations.

• International Journal of Computer Science & Network Security
• /
• 제23권11호
• /
• pp.67-72
• /
• 2023

In the past decade, Autonomous Vehicle Systems (AVS) have advanced at an exponential rate, particularly due to improvements in artificial intelligence, which have had a significant impact on social as well as road safety and the future of transportation systems. The fusion of light detection and ranging (LiDAR) and camera data in real-time is known to be a crucial process in many applications, such as in autonomous driving, industrial automation and robotics. Especially in the case of autonomous vehicles, the efficient fusion of data from these two types of sensors is important to enabling the depth of objects as well as the classification of objects at short and long distances. This paper presents classification of objects using CNN based vision and Light Detection and Ranging (LIDAR) fusion in autonomous vehicles in the environment. This method is based on convolutional neural network (CNN) and image up sampling theory. By creating a point cloud of LIDAR data up sampling and converting into pixel-level depth information, depth information is connected with Red Green Blue data and fed into a deep CNN. The proposed method can obtain informative feature representation for object classification in autonomous vehicle environment using the integrated vision and LIDAR data. This method is adopted to guarantee both object classification accuracy and minimal loss. Experimental results show the effectiveness and efficiency of presented approach for objects classification.