• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.10a
• /
• pp.705-708
• /
• 2009

This research describes a neural network-based phoneme segmenter for recognizing spontaneous speech. The input of the phoneme segmenter for spontaneous speech is 16th order mel-scaled FFT, normalized frame energy, ratio of energy among 0~3[KHz] band and more than 3[KHz] band. All the features are differences of two consecutive 10 [msec] frame. The main body of the segmenter is single-hidden layer MLP(Multi-Layer Perceptron) with 72 inputs, 20 hidden nodes, and one output node. The segmentation accuracy is 78% with 7.8% insertion.

• The Journal of the Korea Contents Association
• /
• v.11 no.5
• /
• pp.32-40
• /
• 2011

The tracking and analysis of cell activities in time-lapse sequences plays an important role in understanding complex biological processes such as the spread of the tumor, an invasion of the virus, the wound recovery and the cell division. For automatic tracking of cells, the tasks such as the cell detection at each frame, the investigation of the correspondence between cells in previous and current frames, the identification of the cell division and the recognition of new cells must be performed. This paper proposes an automatic cell tracking algorithm. In the first frame, the marker of each cell is extracted using the feature vector obtained by the analysis of cellular regions, and then the watershed algorithm is applied using the extracted markers to produce the cell segmentation. In subsequent frames, the segmentation results of the previous frame are incorporated in the segmentation process for the current frame. A combined criterion of geometric and intensity property of each cell region is used for the proper association between previous and current cells to obtain correct cell tracking. Simulation results show that the proposed method improves the tracking performance compared to the tracking method in Cellprofiler (the software package for automatic analysis of bioimages).

• Annual Conference on Human and Language Technology
• /
• 2018.10a
• /
• pp.352-357
• /
• 2018

본 논문에서는 한국어 프레임넷 분석기를 구축하기 위하여 한국어 프레임넷 데이터를 가공하여 공개하고, 한국어 프레임 분류 및 논항의 의미역 분류 문제를 해결하기 위한 방법을 제안한다. 프레임넷은 단어 단위가 아닌 단어들의 범위로 구성된 범위에 대해 어노테이션된 코퍼스라는 점에 착안하여, 어휘 및 논항의 내부 의미 정보와 외부 의미 정보, 그리고 프레임과 각 의미역들의 임베딩을 학습한 중첩 분할된 양방향 LSTM 모델을 사용하였다. 이를 통해 한국어 프레임 분류에서 72.48%, 논항의 의미역 분류에서 84.08%의 성능을 보였다. 또한 본 연구를 통해 한국어 프레임넷 데이터의 개선 방안을 논의한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2005.11a
• /
• pp.189-192
• /
• 2005

비디오 분할의 목적은 같은 내용들을 가지는 프레임들의 순서를 표현하는 각 샷의 비디오 순서 분할을 위한 것이다. 그리고 색인에 대한 각 샷으로부터 키 프레임을 선택한다. 존재하는 비디오 분할 방법들은 2가지 그룹들로 분류될 수 있다. 먼저 경계값이 할당되어야만 하는 샷 전환 검출(SCD) 접근과 클러스터 수의 사전 지식이 요구되는 클러스터 접근이다. 본 논문에서는 컬러-$x^2$명도 히스토그램 기반 FCM(fuzzy c-means) 클러스터링 알고리즘을 사용하는 비디오 분할 방법을 제안하였다. 이 알고리즘은 앞에서 기술한 2가지 접근의 혼합이다. 그리고 이것은 두 가지 접근들의 결점을 극복하도록 설계 되었다. 실험 결과들은 컬러-$x^2$명도 히스토그램 기반 FCM 클러스링 알고리즘이 강건하고 비디오 시퀀스들의 다양한 형태들에 응용할 수 있다고 제안한다.

• The KIPS Transactions:PartB
• /
• v.8B no.5
• /
• pp.493-500
• /
• 2001

Video scene segmentation is fundamental role for content based video analysis. In this paper, we propose a new region based video scene segmentation method using continuity test for each object region which is segmented by the watershed algorithm for all frames in video data. For this purpose, we first classify video data segments into classes that are the dynamic and static sections according to the object movement rate by comparing the spatial and shape similarity of each region. And then, try to segment each sections by grouping each sections by comparing the neighbor section sections by comparing the neighbor section similarity. Because, this method uses the region which represented on object as a similarity measure, it can segment video scenes efficiently without undesirable fault alarms by illumination and partial changes.

• Journal of the Korea Society of Computer and Information
• /
• v.13 no.1
• /
• pp.109-116
• /
• 2008

Even though many studies on parallel rendering based on PC clusters have been done. most of those did not cope with non-uniform scenes, where locations of 3D models are biased. In this work. we have built a PC cluster system with POV-Ray, a free rendering software on the public domain, and developed an adaptive load balancing scheme to optimize the parallel efficiency Especially, we noticed that a frame of 3D animation are closely coherent with adjacent frames. and thus we could estimate distribution of computation amount, based on the computation time of previous frame. The experimental results with 2 real animation data show that the proposed scheme reduces by 40% of execution time compared to the simple static partitioning scheme.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.7
• /
• pp.27-32
• /
• 2010

We can cut and paste portions of MPEG coded bitstream efficiently to rearrange the audio and video sequences using our proposed method. The proposed method decodes the MPEG stream within just only one GOP(Group of Picture), edits the decoded video frames, and encodes it back to a MPEG stream. In this method, precise editing is possible. A pre-processing step is specially designed to provide easy cut and paste processing. In the pre-processing step for editing MPEG streams, the detail information is extracted. In addition, video quality is not degraded after the proposed editing process is applied. Consequently, the experimental results show significant improvements compared with traditional algorithms for video editing method in terms of the efficiency and exactness.

• Journal of KIISE:Software and Applications
• /
• v.36 no.1
• /
• pp.54-61
• /
• 2009

This paper proposes a skin segmentation method based on region histograms of the color quantization map. First, we make a quantization map of the image using the JSEG algorithm and detect the skin pixel. For the skin region detection, the similar neighboring regions are set by its similarity of the size and location between the previous frame and the present frame from the each region of the color quantization map. Then we compare the similarity of histogram between the color distributions of each quantized region and the skin color model using the histogram distance. We select the skin region by the threshold value calculated automatically. The skin model is updated by the skin color information from the selected result. The proposed algorithm was compared with previous algorithms on the ECHO database and the continuous images captured under time varying illumination for adaptation test. Our approach shows better performance than previous approaches on skin color segmentation and adaptation to varying illumination.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2004.05a
• /
• pp.211-214
• /
• 2004

본 논문은 논리 연산을 이용한 실시간 주행 차량 분할 및 추적에 관한 알고리즘을 제안하였다. 연속된 프레임 간에 논리연산을 이용하여 영상을 분할하고, 배경과 잡음을 제거하였으며 영상에서 주행차량의 이동 영역을 추출하였다. 주행차량들을 논리 연산을 이용하여 영상분할 함으로써 기존 방법에 비해 평활화 및 에지추출 단계에서 나타날 수 있는 문제점들을 제거하였고, 전처리 단계를 줄였으며, 알고리즘을 단순화 하였다. 또한 추적되는 영상으로부터 위치와 컬러등의 주행 차량의 특징을 직접 추출 가능하도륵 하였다.

• Journal of KIISE:Computing Practices and Letters
• /
• v.8 no.4
• /
• pp.430-441
• /
• 2002

Video object segmentation is an important component for object-based video coding scheme such as MPEG-4. In this paper, a robust algorithm for segmentation of moving objects in video sequences and VOP(Video Object Planes) extraction is presented. The points of this paper are detection, of an accurate object boundary by associating moving object edge with spatial object edge and generation of VOP. The algorithm begins with the difference between two successive frames. And after extracting difference image, the accurate moving object edge is produced by using the Canny algorithm and morphological operation. To enhance extracting performance, we app]y the morphological operation to extract more accurate VOP. To be specific, we apply morphological erosion operation to detect only accurate object edges. And moving object edges between two images are generated by adjusting the size of the edges. This paper presents a robust algorithm implementation for fast moving object detection by extracting accurate object boundaries in video sequences.