• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.3C
• /
• pp.279-287
• /
• 2009

In this paper, a hierarchical stereo matching with color information is proposed. To generate an initial disparity map, feature based stereo matching is carried out and to generate a final disparity map, hierarchical stereo matching is carried out. The boundary (edge) region is obtained by segmenting a given image into R, G, B and White components. From the obtained boundary, disparity is extracted. The initial disparity map is generated when the extracted disparity is spread to the surrounding regions by evaluating autocorrelation from each color region. The initial disparity map is used as an initial value for generating the final disparity map. The final disparity map is generated from each color region by changing the size of a block and the search range. 4 test images that are provided by Middlebury stereo vision are used to evaluate the performance of the proposed algorithm objectively. The experiment results show better performance compared to the Graph-cuts and Dynamic Programming methods. In the final disparity map, about 11% of the disparities for the entire image were inaccurate. It was verified that the boundary for the non-contiguous point was clear in the disparity map.

• Proceedings of the Korean Information Science Society Conference
• /
• 2007.10c
• /
• pp.459-464
• /
• 2007

영상기반의 3자원 복원(reconstruction)에 대한 연구가 컴퓨터 성능의 발전과 다양한 영상기반의 복원 알고리즘의 연구로 인해 최근 좋은 결과를 보이고 있으나, 이는 얼굴영역과 같은 목적이 되는 영역이 각 입력영상으로부터 미리 정확하게 추출되어 있다고 가정하기 때문이다. 일반적으로 목적이 되는 영역을 추출하기 위해 차영상이 많이 이용되고 있지만 차영상은 잡음과 구멍(hole)과 같은 오 추출된 영역이 발생하기 때문에 목적이 되는 영역을 3차원으로 복원을 할 때 심각한 오류를 초래할 수 있다. 전경물체(목적이 되는 영역)을 정확하게 추출하기 위해 최근 그래프 컷(graph cut)을 이용한 방법이 다양하게 시도되고 있다. 그래프 컷은 데이터 항(data term)과 스무드 항(smooth term)으로 구성된 에너지 함수를 전역적으로 최소화하는 방법으로 여러 공학적 문제에서 좋은 결과를 보이고 있지만, 에너지 함수의 데이터 항을 설정할 때 필요한 사전정보를 자동으로 얻기가 어렵다. 스테레오 비전의 깊이 정보가 최근 전경 물체 추출을 위한 사전정보로 많이 이용되고 있고 그들의 실험환경에서는 좋은 결과를 보이지만, 3차원 얼굴 복원에서 얼굴의 대부분이 동질의 영역을 가지고 있기 때문에 깊이 정보를 구하기 어려워 정확한 사전정보를 구하기가 어렵다. 본 논문에서는 3차원 얼굴 복원을 효과적으로 하기 위한 그래프 컷 기반의 전경 물체 추출 방법을 제안한다. 에너지 함수의 데이터 항을 설정하기 위해 전경 물체에 대한 사전정보를 추출해야 하며, 이를 위해 차영상을 이용하여 대략적인 전경 물체 추출하고, 사전정보에 대한 오류를 줄이기 위해 잡음과 그림자 영역을 제거한다. 잡음과 그림자 영역을 제거하면 구멍이 발생하거나 실루엣이 손상되는 문제가 발생한다. 손상된 정보는 근접한 픽셀이 유사하지 않을 때 낮은 비용을 할당하는 에너지 함수의 스무드(smooth) 항에 의해 에지 정보를 기반으로 채워진다. 결론적으로 제안된 방법은 스무드 항과 대략적으로 설정된 데이터 항으로 구성된 에너지 함수를 그래프 컷으로 전역적으로 최소화함으로써 더욱 정확하게 목적이 되는 영역을 추출할 수 있다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2006.11a
• /
• pp.127-132
• /
• 2006

Quantization in video coding plays an important role in controlling the bit-rate of compressed video bit-streams. It has been used as an important control means to adjust the amount of bit-streams to at]owed bandwidth of delivery networks and storage. Due to the dependent nature of video coding, dependent quantization has been proposed and applied for MPEG-2 video coding to better maintain the quality of reconstructed frame for given constraints of target bit-rate. Since Scalable Video Coding (SVC) being currently standardized exhibits highly dependent coding nature not only between frames but also lower and higher scalability layers where the dependent quantization can be effectively applied, in this paper, we propose a dependent quantization scheme for SVC and compare its performance in visual qualities and bit-rates with the current JSVM reference software for SVC. The proposed technique exploits the frame dependences within each GOP of SVC scalability layers to formulate dependent quantization. We utilize Lagrange optimization, which is widely accepted in R-D (rate-distortion) based optimization, and construct trellis graph to find the optimal cost path in the trellis by minimizing the R-D cost. The optimal cost path in the trellis graph is the optimal set of quantization parameters (QP) for frames within a GOP. In order to reduce the complexity, we employ pruning procedure using monotonicity property in the trellis optimization and cut the frame dependency into one GOP to decrease dependency depth. The optimal Lagrange multiplier that is used for SVC is equal to H.264/AVC which is also used in the mode prediction of the JSVM reference software. The experimental result shows that the dependent quantization outperforms the current JSVM reference software encoder which actually takes a linear increasing QP in temporal scalability layers. The superiority of the dependent quantization is achieved up to 1.25 dB increment in PSNR values and 20% bits saving for the enhancement layer of SVC.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.42 no.6
• /
• pp.93-100
• /
• 2005

We propose an optimal thresholding method for the voxel coloring in the reconstruction of a 3D shape. Our purposed method is a new approach to resolve the trade-off error of the threshold value on determining the photo-consistency in the conventional method. Optimal thresholding value is decided to compare the photo-consistency of a surface with inside voxel on the optic ray of the center camera. As iterating the process of the vokels, the threshold is approached to the optimal value for the individual surface voxel. And also, graph cut method is reduced to the surface noise on eliminating neighboring voxel. To verify the proposed algorithm, we simulated in the virtual and real environment. It is advantaged to speed up and accuracy of a 3D face reconstruction by applying the methods of optimal threshold and graph as compare with conventional algorithms.

• The Journal of the Korea Contents Association
• /
• v.17 no.10
• /
• pp.567-575
• /
• 2017

In this study, when diagnosis pancreatic cancer by dual time point PET/CT, we propose SUVm 2.52 as the threshold value for performing the dual time point PET/CT exam. The hypothesis of normal distribution was adopted through data conversion of 60 pancreatic diseases. The proposed SUVm2.52 boundary value showed a significance level that could be applied to both 120 and 180 minutes of delay time scan for pancreatic cancer determination (p<0.05). C-value variation shows that delay time 2 hour test is more useful than delay time 3 hour test. When the SUVm 2.52 is set to the boundary value and the double-time point PET/CT exam is performed, the probability of distinguishing cancer from inflammation in the delayed image is 95%. When the delayed test is performed with the proposed boundary value SUVm 2.52, Compared with general PET / CT scans, it is thought that it may be helpful to distinguish pancreatic cancer.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.1
• /
• pp.37-42
• /
• 2024

This paper suggests polynomial time algorithm for cutwidth minimization problem that classified as NP-complete because the polynomial time algorithm to find the optimal solution has been unknown yet. To find the minimum cutwidth CW_f(G)=max_𝜈∈_VCW_f(𝜈)for given graph G=(V,E),m=｜V｜, n=｜E｜, the proposed algorithm divides neighborhood N_G[𝜈_i] of the maximum degree vertex 𝜈_i in graph G into left and right and decides the vertical cut plane with minimum number of edges pass through the vertex 𝜈_i firstly. Then, we split the left and right N_G[𝜈_i] into horizontal sections with minimum pass through edges. Secondly, the inner-section vertices are connected into line graph and the inter-section lines are connected by one line layout. Finally, we perform the optimization process in order to obtain the minimum cutwidth using vertex moving method. Though the proposed algorithm requires O(n²) time complexity, that can be obtains the optimal solutions for all of various experimental data

• Fibers and Polymers
• /
• v.7 no.4
• /
• pp.404-413
• /
• 2006

This study examines multiple quality optimization of the injection molding for Polyether Ether Ketone (PEEK). It also looks into the dimensional deviation and strength of screws that are reduced and improved for the molding quality, respectively. This study applies the Taguchi method to cut down on the number of experiments and combines grey relational analysis to determine the optimal processing parameters for multiple quality characteristics. The quality characteristics of this experiment are the screws' outer diameter, tensile strength and twisting strength. First, one should determine the processing parameters that may affect the injection molding with the $L_{18}(2^1{\times}3^7)$ orthogonal, including mold temperature, pre-plasticity amount, injection pressure, injection speed, screw speed, packing pressure, packing time and cooling time. Then, the grey relational analysis, whose response table and response graph indicate the optimum processing parameters for multiple quality characteristics, is applied to resolve this drawback. The Taguchi method only takes a single quality characteristic into consideration. Finally, a processing parameter prediction system is established by using the back-propagation neural network. The percentage errors all fall within 2%, between the predicted values and the target values. This reveals that the prediction system established in this study produces excellent results.

• Journal of IKEEE
• /
• v.13 no.4
• /
• pp.1-7
• /
• 2009

A video segmentation algorithm, which can extract objects even with non-stationary backgrounds, is proposed in this work. The proposed algorithm is composed of three steps. First, we perform an initial segmentation interactively to build the probability density functions of colors per each macro block via kernel density estimation. Then, for each subsequent frame, we construct a coherence strip, which is likely to contain the object contour, by exploiting spatio-temporal correlations. Finally, we perform the segmentation by minimizing an energy function composed of color, coherence, and smoothness terms. Experimental results on various test sequences show that the proposed algorithm provides accurate segmentation results.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2010.07a
• /
• pp.408-410
• /
• 2010

본 논문에서는 새로운 척도에 기반한 다노출 영상 융합 방법을 제안한다. 제안하는 방법에서는 각 노출정도에 따른 픽셀값의 그래프가 컨벡스 형태를 갖는다는 성질과 대조 값의 차이를 고려한 MRF (Markov Random Field) 기반의 에너지 함수를 설계하고 그 에너지 함수를 그래프컷 (Graph cut) 으로 풀어 각 노출치 영상에 대한 가중치 맵 (weight map)을 형성한다. 그리고 가중치 맵을 곱한 각 영상을 더함으로써 융합된 영상을 얻는다. 제안한 컨벡스 성질을 기반으로 한 척도는 특정 컬러 성분이 다른 컬러 성분보다 먼저 과노출 상태에 도달 한 경우의 영역을 가중치 계산에서 제외할 수 있기 때문에 기존의 가중치 기반의 방법보다 정확한 가중치 맵을 형성할 수 있게 한다. 실험 결과 제안하는 방법은 기존의 다노출 영상 융합에 비해 보다 넓은 영역에서 원 영상의 정보를 더 잘 표현하는 것을 확인하였다.

• Journal of the Korea Society of Computer and Information
• /
• v.21 no.9
• /
• pp.79-84
• /
• 2016

In this paper we consider the problem of finding the triplet (S,${\pi}$,f), where $S{\subseteq}V$, ${\pi}$ is a sequence of nodes in S and $f:V{\backslash}S{\rightarrow}S$ for a given complete graph G=(V,E). In particular, there exist two costs, $c^V_{uv}$ and $c^D_{uv}$ for $(u,v){\in}E$, and the cost of triplet (S,${\pi}$,f) is defined as $\sum_{i=1}^{{\mid}S{\mid}}c^V_{{\pi}(i){\pi}(i+1)}+2$ ${\sum_{u{\in}V{\backslash}S}c^D_{uf(u)}$. This problem is motivated by the integrated routing of the vehicle and drone for urban delivery services. Since a well-known NP-complete TSP (Traveling Salesman Problem) is a special case of our problem, we cannot expect to have any polynomial-time algorithm unless P=NP. Furthermore, for practical purposes, we may not rely on time-exhaustive enumeration method such as branch-and-bound and branch-and-cut. This paper suggests the simple heuristic which is motivated by the MST (minimum spanning tree)-based approximation algorithm and neighborhood search heuristic for TSP.