• The Magazine of the IEIE
• /
• v.39 no.2
• /
• pp.45-54
• /
• 2012

Object detection is a fundamental task for many high-level computer vision applications such as image retrieval, scene understanding, activity recognition, visual surveillance and many others. Although object detection is one of the most popular problems in computer vision and various algorithms have been proposed thus far, it is also notoriously difficult, mainly due to lack of proper models for object representation, that handle large variations of object structure and appearance. In this article, we review a branch of object detection algorithms based on Support Vector Machines (SVMs), a well-known max-margin technique to minimize classification error. We introduce a few variations of SVMs-Structural SVMs and Latent SVMs-and discuss their applications to object detection and localization.

• 한국IT서비스학회:학술대회논문집
• /
• 2008.05a
• /
• pp.419-424
• /
• 2008

최근 네트워크 및 정보통신 기술의 발달로 유비쿼터스(Ubiquitous) 환경이 구체화되면서 미래의 정보 기술 산업을 주도할 유비쿼터스 센서 네트워크(Ubiquitous Sensor Network : USN)가 주목받고 있다. 이러한 USN기반의 서비스는 각종 센서들로부터 인식된 데이터를 수집하고 이를 가공하여 사용자가 이해하기 쉽게 USN 자원에 대한 정보를 전달해야 한다. 그러나 USN자원에 대한 정보를 사용자가 이해하기 쉽도록 웹 상에서 시각적으로 표현하기 위한 연구는 보이지 않고 있다. 본 논문에서는 USN 데이터를 크게 센싱데이터와 메타데이터로 분류하고 이를 효과적으로 표현할 수 있는 시각적 요소들을 제안하였다. 그리고 사용자 질의에 대한 응답을 분석하여 각 응답 유형에 따라 시각적 요소를 효과적으로 조합하여 사용자가 쉽게 인지할 수 있는 웹 페이지를 설계 및 구현하였다.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2005.10a
• /
• pp.632-636
• /
• 2005

Design process model represents how a design project proceeds. It encompasses the individual activities of design, their precedence relationships, and the relevant information related to each activity. In contrast to the conventional visual representation methods, ontology-based process model is machine-readable, and therefore it can be implemented in a software system without repeating the whole steps of coding, compiling and link. This paper proposes a framework for design process ontology that defines the relevant objects and attributes in the design process as well as the relationships between them. An example for injection mold design process is shown to explain the substance of the design process model.

• The Journal of Korea Robotics Society
• /
• v.14 no.1
• /
• pp.8-13
• /
• 2019

Visual place recognition is widely researched area in robotics, as it is one of the elemental requirements for autonomous navigation, simultaneous localization and mapping for mobile robots. However, place recognition in changing environment is a challenging problem since a same place look different according to the time, weather, and seasons. This paper presents a feature extraction method using a deep convolutional auto-encoder to recognize places under severe appearance changes. Given database and query image sequences from different environments, the convolutional auto-encoder is trained to predict the images of the desired environment. The training process is performed by minimizing the loss function between the predicted image and the desired image. After finishing the training process, the encoding part of the structure transforms an input image to a low dimensional latent representation, and it can be used as a condition-invariant feature for recognizing places in changing environment. Experiments were conducted to prove the effective of the proposed method, and the results showed that our method outperformed than existing methods.

• Journal of the Korea Society of Computer and Information
• /
• v.24 no.1
• /
• pp.33-39
• /
• 2019

In this paper, we propose deep convolutional neural network(CNN) with bottleneck structure which improves the performance of earthquake classification. In order to address all possible forms of earthquakes including micro-earthquakes and artificial-earthquakes as well as large earthquakes, we need a representation and classifier that can effectively discriminate seismic waveforms in adverse conditions. In particular, to robustly classify seismic waveforms even in low snr, a deep CNN with 1x1 convolution bottleneck structure is proposed in raw seismic waveforms. The representative experimental results show that the proposed method is effective for noisy seismic waveforms and outperforms the previous state-of-the art methods on domestic earthquake database.

• International Journal of Advanced Culture Technology
• /
• v.9 no.1
• /
• pp.105-113
• /
• 2021

Through the analysis of Gerhard Richter's works of art in the period of "fuzzy image", this paper expounds the special internal quality of fuzzy image in Richter's works and reveals an important direction of the development of contemporary easel painting, The special essence of Richter's vague image is that he uses photos to reflect the power and authenticity of the existence of the objective things. He was not satisfied with the radical way of modern painting and tried to return to the traditional way of painting, but in fact Richter kept a special distance between classical painting and modern art. Richter not only blurs the image in his creation, but also shows that he wants to show the objectivity of things, keep the distance, authenticity and give up painting. Richter used fuzzy images to capture the concept of "visual unconsciousness", and finally separated from the concept. When we read how Richter showed the reality of things in the way of painting, we also saw a kind of contradictory psychology when he faced the complicated objective.

• Asian Journal for Public Opinion Research
• /
• v.8 no.4
• /
• pp.409-431
• /
• 2020

This study explores the images and ideas presented by the editorial cartoons that have appeared in the author's Facebook timeline during Duterte's enhanced community quarantine (ECQ). The study analyzed 70 editorial cartoons posted between March 14, 2020, when Duterte declared ECQ in the National Capital Region of the Philippines, and June 22, 2020, a few days before the emergency powers of Duterte expired. This study used (visual-verbal) textual analysis as the research method to surface discourses embedded in the selected editorial cartoons. The editorial cartoons were clustered according to the roles the powerful people play in the images, and the details of each image were compared and contrasted to surface nuances in representation. The 70 editorial cartoons were classified into seven categories: (1) invisible (non-suffering) persons, (2) front liners, (3) privileged homeowners, (4) priority clients, (5) judges, (6) gatekeepers, and (7) dysfunctional public officials. They gravitated towards the tragic realities that call for acts of social justice and equity, and underscore specific contexts that need to be fixed by those in power.

• Journal of Cardiovascular Imaging
• /
• v.31 no.1
• /
• pp.1-17
• /
• 2023

The Coronary Artery Calcium Data and Reporting System (CAC-DRS) is a standardized reporting method for calcium scoring on computed tomography. CAC-DRS is applied on a per-patient basis and represents the total calcium score with the number of vessels involved. There are 4 risk categories ranging from CAC-DRS 0 to CAC-DRS 3. CAC-DRS also provides risk prediction and treatment recommendations for each category. The main strengths of CAC-DRS include a detailed and meaningful representation of CAC, improved communication between physicians, risk stratification, appropriate treatment recommendations, and uniform data collection, which provides a framework for education and research. The major limitations of CAC-DRS include a few missing components, an overly simple visual approach without any standard reference, and treatment recommendations lacking a basis in clinical trials. This consistent yet straightforward method has the potential to systemize CAC scoring in both gated and non-gated scans.

• Journal of applied mathematics & informatics
• /
• v.42 no.2
• /
• pp.433-444
• /
• 2024

Our study explores the connection between the Pythagorean theorem and the Fixed-point theorem in metric spaces. Both of which center around the concepts of distance transformations and point relationships. The Pythagorean theorem deals with right triangles in Euclidean space, emphasizing distances between points. In contrast, fixed-point theorems pertain to the points that remain unchanged under specific transformations thereby preserving distances. The article delves into the intrinsic correlation between these concepts and presents a novel study in Euclidean metric spaces, examining the relationship between contraction mapping and Pythagorean Right Triangles. Practical applications are also discussed particularly in the context of image compression. Here, the integration of the Pythagorean right triangle paradigm with contraction mappings results in efficient data representation and the preservation of visual data relation-ships. This illustrates the practical utility of seemingly abstract theories in addressing real-world challenges.

• Advances in materials Research
• /
• v.13 no.3
• /
• pp.203-210
• /
• 2024

Science mapping is a visual representation of the structure and dynamics of scholarly knowledge. Gas adsorption on catalyst supports is a crucial process in many catalytic reactions. The R package "Bibliometrix" and VosViewer software were employed for science mapping analysis. The results show that the upward trend but fluctuates from year to year for both annual scientific production and average article citations per year. Co-occurrence of the keywords were used to identify the primary fields of study and to map the existing state of research. Trending topics reveal some interesting features that support the growth of research in this field and are associated with emerging disciplines or areas of study that have not been extensively explored.