Earticle

이 연구의 목적은 최근 논의되고 있는 e스포츠 사례들 중 VR-e스포츠의 가능성을 시각현상에서 나타나는 알고리즘 분석을 통해 제시하였다. 최근에 등장하는 실감형 미디어 사례들을 대상으로 시각 알고리즘 분석을 통해 비교분석 하 였다. 연구결과는 다음과 같다. 첫째, e스포츠가 일반 ‘스포츠’로 인정될 수 있는 명확한 기준이 제시되었다. 둘째, e스포츠보다 더 일반 스포츠로써 인정될 수 있는 VR-e스포츠의 가능성을 증명하였다. 셋째, 시각체계분석, 프레임의 유형화, 알고리즘 분석 등을 통해 e스포츠를 넘어 VR-e스포츠로서의 확장성은 물론 학문적인 근거를 제시할 수 있었 다. 연구방법으로는 전통적인 시각체계를 통한 시각 유형화를 구축하고 e스포츠 사례와 실감형 미디어 사례들을 비교 하여 그 결과 값을 시각 알고리즘화 하였다. 최근까지 e스포츠는 산업적 분야의 이익에 관한 연구는 많이 축적 되었 었다. 하지만 이론적 연구는 미흡한 실정이다. 따라서 본 연구를 통해 e스포츠 연구의 방향을 학적 체계와 학적 가치 를 규명하는 연구를 넘어 VR과의 확장성을 찾는데 연구의 초점을 맞추고자 하였다.

The purpose of this study is to present the possibility of VR-eSports among eSports cases currently being discussed through algorithmic analysis that appears in visual phenomena. The recently appeared immersive media cases were compared and analyzed through visual algorithm analysis. The research results are as follows. First, a clear criterion for eSports to be recognized as a general 'sport' was presented. Second, It suggested the possibility of VR-eSports that can be recognized as more general sports than eSports. Third, Through visual system analysis, frame type, and algorithm analysis, it was possible to present academic grounds as well as expandability as VR-eSports beyond eSports. As a research method, visual typification was built through the existing visual system and the results were transformed into visual algorithms by comparing e-sports cases and immersive media cases. Until recently, e-sports had accumulated a lot of research on the interests of the industrial field. However, the theoretical study is insufficient. Therefore, through this study, the direction of eSports research aims to focus on finding scalability with VR beyond the study of identifying the academic system and academic value.

An advanced visual tracking and stable grasping algorithm for a moving object is proposed. The stable grasping points for a moving 2D polygonal object are obtained through the visual tracking system with the Kalman filter and image prediction technique. The accuracy and efficiency are improved more than any other prediction algorithms for the tracking of an object. In the processing of a visual tracking. the shape predictors construct the parameterized family and grasp planner find the grasping points of unknown object through the geometric properties of the parameterized family. This algorithm conducts a process of ‘stable grasping and real time tracking’.

In collective housing, the view quality is considered important in the respect of dwelling quality, economic value, and so on. Due to the densification of collective housing, it is becoming more important to solve problems such as visual infringement caused by interference between adjacent buildings. The purpose of this study is to propose the quantitative analysis methodology the visual openness of individual units in collective housing, and to propose the visual optimization model that derives the optimized collective housing layout. Through the review of the preceding literature, the definition of the view and the standard of the viewing point were prepared, and the visual analysis methodology was established by applying the isovist theory to quantify visual openness. For optimization of view quality, the visual optimization model was established by applying the NSGA-II theory, a kind of genetic algorithm. In order to examine the effectiveness of the visual optimization model, the model was applied to an existing collective housing and alternative layouts were generated as visually optimized. Finally, suitability of the layouts was reviewed.

In complicated scene, in order to solve the temporal occlusion problem of target tracking, a novel particle filter tracking algorithm based on graphical model is proposed. Graphical model is applied to particle filtering in this method. Firstly, dividing the target into several key regions, and extracting the characteristic value of each region. Then, these regions are applied to construct graphical model. In the process of target tracking by using particle filter method, graphical models can compensate for the lacking information of the occluded region. The state of the occluded part can be inferred by the graphical model. Finally, experimental results have demonstrated that the proposed tracking algorithm is effective, and it can reliably track moving target.

Based on the analysis of the shortcomings of the current KLT (Kanade-Lucas-Tomasi) algorithm, a new, self-adaptive tracking algorithm is devised. Through introducing a kind of filtering mechanism, interference from surrounding noise and light on the tracked target is reduced. In order to reduce the error in tracking, a method based on forward-backward error approximation is utilized. Since such approximation reduces the number of visual feature points on the target, when the target changes in its shape tracking failure may easily result. In order to prevent this possible failure, a mechanism that appends the feature points is introduced. Experimental comparisons show that the said algorithm is clearly more effective than other similar algorithms.

In this paper, we propose a novel algorithm to deal with the problem of visual tracking in some challenging situations, which is based on sparse representation and multi-scale block. To build target templates, we select distinguishable features between the target and background in each frame of video sequences, dictionary is built by the multi-scale block of target templates. Then, particle filter generates filter distribution in the next frame, the moving target is framed by affine transformation. To describe the current state of the target, we calculate posterior probability for each particle. Finally, the templates are updated online. The experimental results show that the proposed algorithm is superior in accuracy than the classical tracking algorithm, and it has better robustness against to the target posture changes, partial occlusion and illumination variations.

In this paper, we propose a blind watermarking scheme based on dual transform domains of discrete wavelet transform and discrete fractional random transform. In order to achieve the robustness and security, we also use the visual secret sharing scheme to split the secret image into two shares and use hologram quantization to spread the watermark information and analyze the cover image detail. For the purpose of widely practical application, we use a QR code for the watermark information. The QR code is decomposed into two shares, the first share is generated by the random seed, and the second share is generated with the help of secret image and the first share. Firstly, the two shares are respectively transformed into a hologram by using the hologram quantization, and then a discrete fractional random transform is applied to them. The cover image is decomposed by two-dimension discrete wavelet transform. The subband generated watermark then is embedded into the intermediate frequency components of the cover image. The watermark extraction process is the reverse of the embedding process. Because of the spectrum characteristic and tear resistant of the hologram, the algorithm has good invisibility. The experimental results show that the proposed algorithm is effective and robust against JPEG loss compression, cropping, rotation and noise.

The moving picture experts group (MPEG) immersive video (MIV) technology has been actively developed and standardized to efficiently deliver immersive video to viewers in order for them to experience immersion and realism in various realistic and virtual environments. Such services are provided by MIV technology, which uses multiview videos as input. The pruning process, which is an important component of MIV technology, reduces interview redundancy in multiviews videos. The primary aim of the pruning process is to reduce the amount of data that available video codec must handle. In this study, two approaches are presented to improve the existing pruning algorithm. The first method determines the order in which images are pruned. The amount of overlapping region between the source views is then used to determine the pruning order. The second method considers global region-wise color similarity to minimize matching ambiguity when determining the pruning area. The proposed methods are evaluated under common test condition of MIV, and the results show that incorporating the proposed methods can improve both objective and subjective quality.

In this paper, we present autonomous mobile robot system, which has an image recognition module and perform tasks by itself with manipulator. Autonomous mobile robot can push each individual button on an elevator panel from a person's command. After the robot receives its command, it analyzes the image information of the elevator button that is acquired by CCD camera in an indoor environment with an elevator. In order to recognize the numbers on the button, the robot separates the number area in button and recognizes the segment through a neural network algorithm. We use the bar code form on the manipulator to find the position of the end of manipulator. The validity of the proposed method i...

4th industrial age! 5G-based video content has changed rapidly. How do you view and understand VR/AR/MR/holograms in the age of convergent technology media represented by digital convergence in the 21st century beyond the modern visual system of Descartes? It predicts extension to an interactive visual system. This study proposes a methodology to visualize the Lacanian visual acuity after applying it to the L-System and to understand the visual system of the 21st century.

We propose an attention based object recognition system, to recognize object fast and robustly. For this we calculate visual stimulus degrees and make saliency maps. Through this map we find a strongly attentive part of image by stimulus degrees, where local features are extracted to recognize objects.

This paper deals with the detection, feature extraction and classification of surface defects in cold rolled strips. Inspection systems are one of the most important fields in factory automation. Defects such as slipmark and dullmark can be effectively detected with a Gaussian matched filter because their shapes are similar to Gaussian. It is justified that the proposed WF(Wavelet Frame) method could be regarded as multiscale Gaussian matched filter which can be applied to the inspection of cold rolled strip. After a wavelet frame transform, the entropies and moments are computed for each subband which pass through both local low pass filter and nonlinear operator. With these features as input, a MLP(Multi Layer Perceptron) is used as a classifier. The proposed inspection method was applied to the real images with defects, and hence showed good performance. The role of each extracted feature is analyzed by KLT(Karhunen-Loeve Transform).

In this paper, we developed visual tracking algorithm using double active bar. The active bar model to represent the object can reduce the search space of energy surface and better performance than those of snake model. However, the contour will not find global equilibrium when driving force caused by image may be weak. To overcome this problem. Double active bar is proposed for finding the global minimum point without any dependence on initialization. To achieve the goal, an deformable model with two initial contours in attempted to search for a global minimum within two specific initial contours. This approach improve the performance of finding the contour of target. To evaluate the performance, some experiments are executed. We can achieved the good result for tracking a object on noisy image.

In this paper, we consider the problems of tracking an object in a real image. In evaluating these problems, we explore a new technique based on an active contour model commonly called a snake model, and propose the active bar models to represent target. Using this model, we simplified the target welection problems, reduced the search space of energy surface, and obtained the better performances than those of snake model. This approach improves the numerical stability and the tendency for points to bunch up and speed up the computational efficiency. Representing the object by active bar, we can easily obtain the zeroth, the first, and the second moment and it facilitates the target tracking. Finally, we present the good result for the visual tracking problem.

One of the challenging tasks in visual inspection using CCD camera is to identify surface defects in an image with complex textured backgeound. In microscopic view, the surface of real objects shows regular or random textured patterns. In this paper, we present a visual inspection algorithm to extract abnormal surface defects in an image with textured background. The algorithm uses the space and frequency information at the same time by introducing the Dominant Frequency Map(DFM) which can describe the frequency characteristics of every small local region of an input image. We demonstrate the feasibility and effectiveness of the method through a series of real experiments for a 14" TV CRT mold. The method successfully identifies a variety of infinitesimal defects, whose size is larger than $50\mu\textrm{m}$, of the mold. The experimental results show that the DFM based method is less sensitive to the environmental changes, such as illumination and defocusing, than conventional vision techniques.ques.

본 논문에서는 옥외 환경에 강인한 영상 감시알고리듬을 구현하는 과정을 기술하였다. 옥외 감시시스템의 어려운 처리 과정들 중 하나는 배경화면을 효과적으로 갱신하는 것이다. 배경 영상에는 건물, 나무들, 이동하는 구름 및 기타 다른 물체들의 그림자를 포함하기 때문에. 시간과 조명광에 따라 변화한다. 이는 옥외에서의 감시시스템의 성능을 저하시킨다. 따라서 본 논문에서는 배경 영상을 효과적으로 갱신하기 위해 적응 혼합 가우시안 필터와 컬러불변성을 화소레벨에서 적용하여 옥외에서도 강인한 영상 감시알고리듬을 제안하였다. 그 결과, 다양한 그림자가 있는 옥외에서 움직이는 대상 물체를 검출할 수 있음을 확인하였다.

This paper describes a robust visual surveillance algorithm under outdoor environment. One of the difficult problems for outdoor is to obtain effective updating process of background images. Because background images generally contain the shadows of buildings, trees, moving clouds and other objects, they are changed by lapse of time and variation of illumination. They provide the lowering of performance for surveillance system under outdoor. In this paper, a robust algorithm for visual surveillance system under outdoor is proposed, which apply the mixture Gaussian filter and color invariant property on pixel level to update background images. In results, it was showed that the moving objects can be detected on various shadows under outdoor.

This paper proposes an advanced visual tracking algorithm for the stable grasping of a moving target(2D). This algorithm is programmed to find grasping points of an unknown polygonal object and execute visual tracking. The Kalman Filter(KF) algorithm based on the SVD(Singular Value Decomposition) is applied to the visual tracking system for the tracking of a moving object. The KF based on the SVD improves the accuracy of the tracking and the robustness in the estimation of state variables and noise statistics. In addition, it does not have the numerical unstability problem that can occur in the visual tracking system based on Kalman filter. In the grasping system, a parameterized family is constructcd, and through the family, the grasping system finds the stable grasping points of an unknown object through the geometric properties of the parameterized family. In the previous studies, many researchers have been studied on only 'How to track a moving target'. This paper concern not only on 'how to track' but also 'how to grasp' and apply the grasping theory to a visual tracking system.

본 논문에서는 옥외 환경에 강인한 영상 감시알고리듬을 구현하는 과정을 기술하였다. 옥외 감시시스템의 어려운 처리 과정들 중 하나는 배경화면을 효과적으로 갱신하는 것이다. 배경 영상에는 건물, 나무들, 이동하는 구름 및 기타 다른 물체들의 그림자를 포함하기 때문에. 시간과 조명광에 따라 변화한다. 이는 옥외에서의 감시시스템의 성능을 저하시킨다. 따라서 본 논문에서는 배경 영상을 효과적으로 갱신하기 위해 적응 혼합 가우시안 필터와 컬러불변성을 화소레벨에서 적용하여 옥외에서도 강인한 영상 감시알고리듬을 제안하였다. 그 결과, 다양한 그림자가 있는 옥외에서 움직이는 대상 물체를 검출할 수 있음을 확인하였다.

This paper describes a robust visual surveillance algorithm under outdoor environment. One of the difficult problems for outdoor is to obtain effective updating process of background images. Because background images generally contain the shadows of buildings, trees, moving clouds and other objects, they are changed by lapse of time and variation of illumination. They provide the lowering of performance for surveillance system under outdoor. In this paper, a robust algorithm for visual surveillance system under outdoor is proposed, which apply the mixture Gaussian filter and color invariant property on pixel level to update background images. In results, it was showed that the moving objects can be detected on various shadows under outdoor.

최근 멀티미디어 콘텐츠(영화, 음악 등)를 스마트폰이나 태블릿 PC, 랩탑 등의 기기에서 끊김 없이 연속적으로 받을 수 있는 N 스크린에 대한 관심이 커지고 있다. 기존에도 화질에 대한 향상을 위한 부선 자원 관리 문제를 푼 경우는 있었지만 다양한 기기의 해상도를 고려하는 N스크린 서비스의 화질 불균형 문제를 해결하고자 하는 노력은 부족했다. 본 논문은 N스크린 멀티캐스트 환경에서 기기의 해상도와 각 그룹마다 시용자의 만족도를 고려한 다중 레이어의 중요도를 새롭게 정의하고, 정의된 중요도에 따라 각 멀티캐스트 그룹의 통신 대역폭과 모듈레이션을 조정해 사용자의 만족도를 최대화하는 과정을 소개하고 있다. 또한 시뮬레이션 과정을 통해서 시각적 중요도를 가장 많이 증가시킬 수 있는 방법으로 자원을 할당하는 것이 실제로 멀티캐스트를 그룹 구성원들에게 화질의 이득을 가져오는 것을 확인하였다.