Siwoo Byun

Asymmetric Index Management Scheme for High-capacity Compressed Databases

Traditional databases exploit a record-based model, where the attributes of a record are placed contiguously in a slow hard disk to achieve high performance. On the other hand, for read-intensive data analysis systems, the column-based compressed database has become a proper model because of its superior read performance. Currently, flash memory SSD is largely recognized as the preferred storage media for high-speed analysis systems. This paper introduces a compressed column-storage model and proposes a new index and its data management scheme for a high-capacity data warehouse system. The proposed index management scheme is based on the asymmetric index duplication and achieves superior search performance using the master index and compact index, particularly for large read-mostly databases. In addition, the data management scheme contributes to the read performance and high reliability by compressing the related columns and replicating them in two mirrored SSD. Based on the results of the performance evaluation under the high workload conditions, the data management scheme outperforms the traditional scheme in terms of the search throughput and response time.

A Column-Aware Index Management Using Flash Memory for Read-Intensive Databases

Abstract Most traditional database systems exploit a record-oriented model where the attributes of a record are placed contiguously in a hard disk to achieve high performance writes. However, for read-mostly data warehouse systems, the column-oriented database has become a proper model because of its superior read performance. Today, flash memory is largely recognized as the preferred storage media for high-speed database systems. In this paper, we introduce a column-oriented database model based on flash memory and then propose a new column-aware flash indexing scheme for the high-speed column-oriented data warehouse systems. Our index management scheme, which uses an enhanced B + -Tree, achieves superior search performance by indexing an embedded segment and packing an unused space in internal and leaf nodes. Based on the performance results of two test databases, we concluded that the column-aware flash index management outperforms the traditional scheme in the respect of the mixed operation throughput and its response time.

Robust Detection of Body Areas Using an Adaboost Algorithm

Recently, harmful content (such as images and photos of nudes) has been widely distributed. Therefore, there have been various studies to detect and filter out such harmful image content. In this paper, we propose a new method using Haar-like features and an AdaBoost algorithm for robustly extracting navel areas in a color image. The suggested algorithm first detects the human nipples through color information, and obtains candidate navel areas with positional information from the extracted nipple areas. The method then selects real navel regions based on filtering using Haar-like features and an AdaBoost algorithm. Experimental results show that the suggested algorithm detects navel areas in color images 1.6 percent more robustly than an existing method. We expect that the suggested navel detection algorithm will be usefully utilized in many application areas related to 2D or 3D harmful content detection and filtering.

Decision of Image Harmfulness Using an Artificial Neural Network

Various types of multimedia contents have been widely spread and distributed with the Internet that is easy to use. Meanwhile, Multimedia contents can bright a social problem because juveniles can access such harmful contents easily through the Internet. This paper proposes a method to determine if an input image is harmful or not, using an neural network. The proposed method first detects a face region from an input image through MCT features. The method then extracts skin color regions using color features and obtains candidate nipple areas from the extracted skin regions. Subsequently, we determine if the input image is harmful, by filtering out non-nipple regions using the artificial neural network. Experimental results show that the proposed method can effectively determine the harmfulness of input images.

Clustering-Based Pattern Abnormality Detection in Distributed Sensor Networks

We suggest a method of effectively detecting and classifying network traffic attacks by visualizing their IP (Internet protocol) addresses and ports and clustering the visualized ports based on their variance. The proposed approach first visualizes the IP addresses and ports of the senders and receivers into two-dimensional images. The method then analyzes the visualized images and extracts linear and/or high brightness patterns, which represent traffic attacks. Subsequently, it groups the ports using an improved clustering algorithm, allowing an artificial neural network to learn the extracted features and to automatically detect and classify normal traffic data, DDoS attacks, DoS attacks, or Internet Worms. The experiments conducted in this work prove that our suggested clustering-based algorithm effectively detects various traffic attacks.

Mirror-Switching Scheme for High-Speed Embedded Storage Systems

The flash memory has been remarked as the next generation media of portable and desktop computers` storage devices. Their features include non-volatility, low power consumption, and fast access time for read operations, which are sufficient to present flash memories as major data storage components for desktop and servers. The purpose of our study is to upgrade a traditional mirroring scheme based on SSD storages due to the relatively slow or freezing characteristics of write operations, as compared to fast read operations. For this work, we propose a new storage management scheme called Memory Mirror-Switching based on traditional mirroring scheme. Our Mirror-Switching scheme improves flash operation performance by switching write-workloads from flash memory to RAM and delaying write operations to avoid freezing. Our test results show that our scheme significantly reduces the write operation delay and storage freezing.

Detection of Special Effects with Circular Moving Borders

Abstract In this paper, we propose a method to detect Iris Round wipe transitions with circular moving borders in digital video data. The suggested method robustly extracts circular moving borders from the input image using improved Hough transform, and finally detects Iris Round wipes by effectively analyzing their moving directions and shapes. In order to evaluate the performance of the suggested algorithm, the experimental results show that the proposed method can effectively detect Iris Rounds with circular moving borders in various video data. Key Words : Special Effect, Circular Moving Border, Video Segmentation, Difference Image * 교신저자 : 변시우(swbyun@anyang.ac.kr)접수일 11년 05월 04일 수정일 11년 06월 07일 게재확정일 11년 07월 07일 1. 서론 비디오 처리 기술들 중에서 장면전환 검출과 같은 비디오 분할은 비디오 인덱싱(indexing)을 위한 첫 번째 단계로, 비디오 시퀀스를 인접한 장면(shot)으로 분할하고 인접한 두 장면을 연결하는 경계를 검출하는 작업이다. 그리고 이러한 장면전환 검출은 자동화되는 것이 매우 중요하다[1].비디오 데이터의 장면전환 검출은 크게 급진적인 장면전환인 컷(cut)과 점진적인 장면전환인 페이드(fade), 디졸브(dissolve), 와이프(wipe) 등으로 분류될 수 있다. 컷은 인접한 영상 사이에서 장면이 급변하는 전환이다. 페이드는 고정 영상(constant image)이 점진적으로 장면화되는 전환인 페이드 인(fade in)과 장면이 점진적으로 고정 영상으로 변화하는 전환인 페이드 아웃(fade out)이 있다. 그리고 디졸브는 페이드 아웃과 페이드 인이 교차해서 발생하는 장면전환이다. 와이프는 새로운 장면이 현재 장면으로 미끄러져 들어오는 점진적인 장면전환으로, 이 전환은 임의의 모양을 가지는 여러 개의 이동 경계선(moving border)을 포함할 수 있다. 와이프 전환은 30가지 종류 이상이 있으며, 경계선의 모양, 이동하는 속도 및 방향에 따라 그 종류가 결정된다[2]. 기존의 장면전환 검출 방법들은 대부분 급진적인 장면전환인 컷 검출을 위주로 수행되고 있으며, 페이드와 디졸브 같은 점진적인 장면전환 검출도 어느 정도는 연구가 진행되었다. 그러나 와이프 장면전환은 다른 전환에 비해서 복잡하고 다양한 변환 패턴 때문에 상대적으로 적은 관심을 받았으며, 기존 연구의 완성도도 상대적으로 낮다. 그럼에도 불구하고 비디오에 존재하는 장면경계의 정확한 검출, 비디오 장르의 인식, 그리고 다른 점진적인