| Title | Target and Background Area Weighted Tracking Algorithm |
| Author | Euncheol Choi, *Young Duk Kim (Yonsei University, Republic of Korea), Suk-Ho Lee (Dongseo University, Republic of Korea), Moon Gi Kang (Yonsei University, Republic of Korea) |
| Page | pp. 741 - 744 |
| Keyword | tracking, centroids shift, mean shift, non-stationary camera |
| Abstract | Recently, kernel-based tracking algorithms such as the mean shift tracking algorithm has been proposed. However, there exists an inherent instability problem which is due to the use of an isotropic kernel for spatiality. In this paper, In this paper, we propose a new tracking algorithm : the weighted mean of the centroids corresponding to each color bin of the target. The weight is determined by the area of the color bins and background colors. The tracking based on proposed model contains spatial information on the distribution of the colors, is rather insensitive to the loss of pixels and change in the number of pixels, and takes the colors into account according to the area they cover in the initial target region. Due to these properties, it possible to track the target in difficult conditions such as low-frame-rate environment, severe partial occlusion and partial color change environment. Furthermore, the target position estimation is executed in a one step computation, which makes the algorithm fast. We compare the stableness of the proposed tracking scheme with the conventional mean shift algorithm experimentally. |
| Title | An Efficient Threshold Method using Gaussian Modeling and Fuzziness Measurement for Moving Target Detection |
| Author | *Jae-Ho Lee (School of Sensor and Display Engineering, Kyungpook National University, Republic of Korea), Ju-Young Kim, Tae-Kyu Kim (School of Electrical Engineering and Computer Science, Kyungpook National University, Republic of Korea), Ki-Hong Kim (Agency for Defense Development, Republic of Korea), Duk-Gyoo Kim (School of Electrical Engineering and Computer Science, Kyungpook National University, Republic of Korea) |
| Page | pp. 745 - 748 |
| Keyword | ATR system, Gaussian modeling, Fuzziness measurement, Target Detection |
| Abstract | In this paper, we propose an efficient algorithm for moving target detection in many military applications such as automatic target recognition (ATR) systems. This algorithm first utilizes a double change detection method (DCDM) that briefly find the motion information of targets. The proposed algorithm employs adaptive threshold method consisted of Gaussian modeling (GM) for obtaining initial threshold and fuzzy information measurement (FIM) method to measure fuzziness degree on adjacent gray levels of the threshold vale of GM method for obtaining optimal threshold value.The results of simulation show that the proposed algorithm has significantly the good detection performance than the conventional methods. |
| Title | Evaluation of 3D Data Service Based on Depth Image Based Rendering over T-DMB |
| Author | *Youngjin Oh, Kwanghee Jung, Joong Kyu Kim (Sungkyunkwan University, Republic of Korea), Gwangsoon Lee, Hyun Lee, Namho Hur, Jinwoong Kim (Electronics and Telecommunications Research Institute, Republic of Korea) |
| Page | pp. 749 - 752 |
| Keyword | Depth-Image-Based-Rendering, 3D-TV, T-DMB, evaluation |
| Abstract | 3D data service over Terrestrial-Digital Multimedia Broadcasting (T-DMB) is very attractive because the single user environment of T-DMB is suitable to glassless 3D viewing. However, the bit budget for transmission of additional 3D data over T-DMB is very limited with 32Kbps through data service channel. To overcome this limited condition, Depth-Image-Based-Rendering (DIBR) can be applied, because corresponding depth sequence is only additionally needed to current T-DMB and this can be compressed effectively. Therefore, in this paper, we evaluate 3D data service based on DIBR over T-DMB. Evaluation mainly consists of two experiments. One is to evaluate the coding efficiency of depth sequences and the other is to measure the subjective quality and percived depth of auto-stereosopic image generated by DIBR with coded depth sequence. Therefore, we evaluate the possibility of transmitted depth image through data service channel that transmission rate is 32Kbps. Evaluation results show that DIBR can efficiently be utilized for 3D data service over T-DMB. However, it is also shown that the development of some techniques such as the depth preprocessing is required for the improvement of image quality. |
| Title | Multi-directional Greedy Stereo Matching |
| Author | *Seung-Hae Baek, Soon-Yong Park, Soon-Ki Jung (Kyungpook National University, Republic of Korea), Sang-Hee Kim (ADD, Republic of Korea), Jeong-Hwan Kim (AR Vision Inc, Republic of Korea) |
| Page | pp. 753 - 756 |
| Keyword | Greedy algorithm, Global stereo matching, RANSAC |
| Abstract | In this paper, we propose a new stereo matching method to reduce a memory size and handle an large image such as satellite images while its performance satisfies speed and accuracy. Our new method is based on a Multi-directional Greedy algorithm and RANSAC. First, we obtain a depth image along multi-directioanl scanlines using the Greedy algorithm. Then, we find reliable areas from the distribution of several depth images using RANSAC. Finally, we extend the reliable areas by iterating the Greedy algorithm and RANSAC several times starting with the previously obtained reliable areas and decide a final depth image. Experimental results show that our algorithm proves the possibility of stereo matching for an enormous image using low memory size while satisfying the performance from the view point of speed and accuracy |
| Title | GLORY-DB: A Distributed Data Management System for Large Scale High-Dimensional Data |
| Author | *Hyun Hwa Choi, Hun Soon Lee, Kyeong Hyeon Park, Mi Young Lee (Electronics and Telecommunications Research Institute, Republic of Korea) |
| Page | pp. 757 - 760 |
| Keyword | cluster system, column-based DBMS, contents-based retrieval, high-dimensional index |
| Abstract | : Recently, the proliferation of the web and digital photography has resulted in the need of a distributed storage system for managing large scale data and an indexing technique for supporting efficient nearest neighbor search on high-dimensional data. One of the most challenging areas in the fields of a distributed data managing and image processing is scalability of data and machines. Especially, for a large scale image clustering problem, which can not fit on a single machine, the traditional nearest neighbor search can not be applied. This paper presents the design of a distributed data management system, highly available and scalable storage system which provides contents-based retrieval using a hybrid spill tree with local signature files. We describe our scalable index structure and how it can be used to find the nearest neighbors in the cluster environments. |