http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Fast Outlier Removal for Image Registration based on Modified K-means Clustering
소영성,Mudasar Qadir,김인택 한국융합신호처리학회 2015 융합신호처리학회 논문지 (JISPS) Vol.16 No.1
Outlier detection and removal is a crucial step needed for various image processing applications such as image registration. Random Sample Consensus (RANSAC) is known to be the best algorithm so far for the outlier detection and removal. However RANSAC requires a cosiderable computation time. To drastically reduce the computation time while preserving the comparable quality, a outlier detection and removal method based on modified K-means is proposed. The original K-means was conducted first for matching point pairs and then cluster merging and member exclusion step are performed in the modification step. We applied the methods to various images with highly repetitive patterns under several geometric distortions and obtained successful results. We compared the proposed method with RANSAC and showed that the proposed method runs 3~10 times faster than RANSAC.
소영성,홍정우 한국융합신호처리학회 2015 융합신호처리학회 논문지 (JISPS) Vol.16 No.3
Connected component labeling (CCL) is a mandatory step in image segmentation where objects are extracted and uniquely labeled. CCL is a computationally expensive operation and thus is often done in parallel processing framework to reduce execution time. Various parallel CCL methods have been proposed in the literature. Among them are NSZ label equivalence (NSZ-LE) method, modified 8 directional label selection (M8DLS) method, HYBRID1 method, and HYBRID2 method. Soh et al. showed that HYBRID2 outperforms the others and is the best so far. In this paper we propose a new hybrid parallel CCL algorithm termed as HYBRID3 that combines selective four directional label search (S4DLS) with label backtracking (LB). We show that the average percentage speedup of the proposed over M8DLS is around 60% more than that of HYBRID2 over M8DLS for various kinds of images.
소영성 明知大學校 産業技術硏究所 1993 産業技術硏究所論文集 Vol.12 No.-
Subbarao and Waxman gave the closed from solution to the nonlinear formulation of motion and structure equation by Waxman and Ullman. They give the rigorous proof on the uniqueness of their closed form optical flow solutions for planar surfaces in motion. However, they fail to consider one special case where the planar patch parallel to the image plane is translating only in Z direction. In this paper, we give a separate closed form solution for this special case, thus completing the solution set for optical flow equations.
Face Tracking based on Track Graph and Hypothesize-and-Verify Formalism in a Video Stream
蘇英聖,鄭燦基 明知大學校 産業技術硏究所 2005 産業技術硏究所論文集 Vol.24 No.-
In this paper, we describe methods for tracking human face in the video taken by a stationary camera. To track face, we first track human body and find face in the body. Conventional methods show weakness in situations where multiple people merge and split in a complicated way. We propose to use track graph where the node contains information of a blob of human body(s) and the link represents temporal correspondences among blobs. To solve heavy merge and split problems, we use color information along with track graph. We also apply hypothesize-and-verify formalism to disambiguate complex incidents where merge and split create ambiguities. To detect face from the result of human body tracking, we use YUV color range, size, and possible locations of face.
An Improved Hybrid Approach to Parallel Connected Component Labeling using CUDA
소영성,Hadi Ashraf,김인택 한국융합신호처리학회 2015 융합신호처리학회 논문지 (JISPS) Vol.16 No.1
In many image processing tasks, connected component labeling (CCL) is performed to extract regions of interest. CCL was usually done in a sequential fashion when image resolution was relatively low and there are small number of input channels. As image resolution gets higher up to HD or Full HD and as the number of input channels increases, sequential CCL is too time-consuming to be used in real time applications. To cope with this situation, parallel CCL framework was introduced where multiple cores are utilized simultaneously. Several parallel CCL methods have been proposed in the literature. Among them are NSZ label equivalence (NSZ-LE) method[1], modified 8 directional label selection (M8DLS) method[2], and HYBRID1 method[3]. Soh [3] showed that HYBRID1 outperforms NSZ-LE and M8DLS, and argued that HYBRID1 is by far the best. In this paper we propose an improved hybrid parallel CCL algorithm termed as HYBRID2 that hybridizes M8DLS with label backtracking (LB) and show that it runs around 20% faster than HYBRID1 for various kinds of images.
소영성,송재현,해용석 한국융합신호처리학회 2018 융합신호처리학회 논문지 (JISPS) Vol.19 No.2
The analysis of plankton species distribution in sea or fresh water is very important in preserving marine ecosystem health. Since manual analysis is infeasible, many automatic approaches were proposed. They usually use images from in situ towed underwater imaging sensor or specially designed, lab mounted microscopic imaging system. Normally they assume that only single plankton is present in an image so that, if there is a clumping among multiple plankton of same species (homogeneous clumping) or if there are multiple plankton of different species scattered in an image (heterogeneous interspersion), they have a difficulty in recognition. In this work, we propose a deep learning based method that can detect and recognize individual plankton in images with homogeneous clumping, heterogeneous interspersion, or combination of both.
동영상 분석 및 이동물체 추적을 위한 새로운 두단계 영역상응 방법
백경민,소영성 明知大學校 産業技術硏究所 1994 産業技術硏究所論文集 Vol.13 No.-
동영상 분석이 정지영상 분석보다 더 많은 3차원 정보를 전달해주기 때문에 컴퓨터 비전 분야에서 많은 주목을 받고있다. 동영상 분석의 대부분 연구자들은 어느 정도의 상응이 이미 해결되었다고 가정을한다. 그러나 상응의 입증은 그 자체가 매우 복잡한 문제이다. 이 논문에서는 두가지 일반적인 방법(거리와 유사도)을 혼합한 두 단계 영역 상응 방법을 제안한다.
양상규,소영성 明知大學校 産業技術硏究所 1996 産業技術硏究所論文集 Vol.15 No.-
Due to the rapid increase of vehicles and poor availability of roads, traffic congestion problem is about to explode. To solve this problem we need real time information about traffic flow to control traffic signals dynamically. Until now loop coil is the most prevalent sensor used for obtaining traffic flow information. However, it is not able to track individual vehicles which is essential in estimating the average vehicle speed. As a result, image sensors started to find their role in this problem domain. Several systems based on image sensors were proposed which assumes either gray level or color image sequence. In this paper, we propose region correspondence based moving vehicle tracking method using color information.
양상규,소영성 明知大學校 産業技術硏究所 1997 産業技術硏究所論文集 Vol.16 No.-
Dynamic scene analysis has been drawn much attention in computer vision field since it reveals more 3-D information than static scene counterpart. The objective of the proposed research is to count number of vehicles that pass through designated detection line in gray-scale image sequence. In this paper, we propose a vehicle counting method using image processing in gray-scale image sequence. We use contiguous frame differencing and morphology along with trip-line method.