http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
- 中文 을 입력하시려면 zhongwen을 입력하시고 space를누르시면됩니다.
- 北京 을 입력하시려면 beijing을 입력하시고 space를 누르시면 됩니다.
RISS 인기검색어
- 검색키워드
- 저자 : Jooyoung Hahn (검색결과 23 건)
- 무료
- 기관 내 무료
- 유료
-
FINE SEGMENTATION USING GEOMETRIC ATTRACTION-DRIVEN FLOW AND EDGE-REGIONS
Jooyoung Hahn,Chang-Ock Lee 한국산업응용수학회 2006 한국산업응용수학회 학술대회 논문집 Vol.1 No.2
In the segmentation problems to extract objects from an image to make, for examples, 3D VR (virtual reality) contents or to estimate sizes of objects, a key issue is fine segmentation which means that the objects can be extracted without visual loss of detailed shapes. Our research is motivated by making 3D VR contents of commercial products. It makes an e-catalog that customers can browse a product in three dimensional virtual space on internet markets. A common way of making a 3D VR content starts from taking hundreds of photographs of a product with different view angles in a photo studio. The most difficult step is to extract the product from a background without visual loss of detailed shapes. The images taken in the studio have wellknown difficulties in segmentation problems even though they usually have simple background colors and small amount of noises such as JPEG artifacts. These mainly come from lighting conditions in the studio and complex shapes of products. Most of lighting conditions make shadows which cause weak boundaries between objects and the background. More serious weak boundaries are produced by a reflection on some parts of an object due to bright lighting conditions and properties of materials of the object. It changes colors of objects into almost white which is normally used as a background color. In addition, there are other difficulties; shapes of objects can be highly non-convex. There have been a lot of boundary-based segmentation algorithms. The snake model in [1] has been a foundation of curve evolution based on the minimization of an energy. The geodesic active contour model was introduced in [2] as the minimization of a weighted length. Although the model has many advantages over the classical snake, it has drawbacks such as dependence on positions of initial curves, incapacity for capturing weak boundaries when an image has both weak and strong boundaries, and slow convergence in non-convex boundaries. Numerous modifications of the snake model and the geodesic active contour model have been developed. to address these drawbacks. In [3], gradient vector flow was proposed for a fast convergence to the non-convex boundaries. In [4], a curvature vector flow was introduced to overcome a limitation of [3] for segmenting highly non-convex shapes. In [5], the region-aided geometric snake was proposed for more robust detection of weak edges. If an object in an image has both weak boundaries and highly non-convex shapes, most of boundary-based segmentation algorithms suffer from capturing such boundaries all around the object. Even though they capture the boundaries, it is not enough to be a fine segmentation for extracting the detailed object from an image. In this talk, we propose a fine segmentation algorithm for extracting objects in an image, which have both weak boundaries and highly non-convex shapes. There are two main concepts, geometric attraction-driven flow (GADF) and edge-regions, which are combined to detect boundaries of objects in a sub-pixel resolution. Since an image is a two dimensional manifold, we obtain GADF by comparing two lengths of curves along the direction of the largest change in the manifold. Edge-regions contain most of edges. We compute inward fluxes in the gradient field of a strength of edges to obtain such regions by which we construct initial curves close to boundaries of objects. Since the orientation of GADF near boundaries of objects points to edges from each side of the boundaries regardless of strength of edges, we can segment the objects by solving a simple advection equation of curves in the flow. It naturally solves problems of a slow convergence near highly non-convex boundaries and a leakage over weak edges. According to the purpose of segmentation, for examples, fine extraction of objects or measurement of sizes of objects, we additionally propose a local region competition algorithm to obtain perceptible boundaries which are used for extractio
-
-
FINE SEGMENTATION USING GEOMETRIC ATTRACTION-DRIVEN FLOW AND EDGE-REGIONS
JOOYOUNG HAHN,CHANG-OCK LEE 한국산업응용수학회 2007 Journal of the Korean Society for Industrial and A Vol.11 No.2
A fine segmentation algorithm is proposed for extracting objects in an image, which have both weak boundaries and highly non-convex shapes. The image has simple background colors or simple object colors. Two concepts, geometric attraction-driven flow (GADF) and edge-regions are combined to detect boundaries of objects in a sub-pixel resolution. The main strategy to segment the boundaries is to construct initial curves close to objects by using edge-regions and then to make a curve evolution in GADF. Since the initial curves are close to objects regardless of shapes, highly non-convex shapes are easily detected and dependence on initial curves in boundary-based segmentation algorithms is naturally removed. Weak boundaries are also detected because the orientation of GADF is obtained regardless of the strength of boundaries. For a fine segmentation, we additionally propose a local region competition algorithm to detect perceptible boundaries which are used for the extraction of objects without visual loss of detailed shapes. We have successfully accomplished the fine segmentation of objects from images taken in the studio and aphids from images of soybean leaves.
-
-
Time Series Simulation of Explosive Charges In Shallow Water Using Ray Approach
Hahn, Jooyoung,Lee, Seongwook,Na, Jungyul The Acoustical Society of Korea 2003 韓國音響學會誌 Vol.22 No.e3
A time series simulation is presented by a ray approach for the simulating the received waveform of a broadband acoustical signals interacting with the ocean boundaries. The environment is assumed to be horizontally stratified, and the seafloor is described in terms of homogeneous fluid half-space. The ray approach includes the effects of reflection from the air-water, water-sediment interface and phase shifts due to boundaries interaction. To generate time series, we assume that the acoustic energy propagates from source to receiver along eigenrays and represent the action of the bottom on the incident wave by a linear filter and characterized in the frequency domain by the transfer function. As example application, the time series for an explosive source in a shallow water environment is calculated and analyzed in terms of acoustical process. good agreement with measured time series is demonstrated.
-
Measurement of bottom-reflected sound in bottom-limited propagation environment
Hahn, Jooyoung,Park, Joungsoo Institute of Pure and Applied Physics 2016 Japanese Journal of Applied Physics Vol.55 No.7
<P>To study the bottom reflection of underwater acoustic sound in a bottom-limited propagation environment, an experiment was conducted using four transmitting sounds in the form of a continuous wave from 1 to 6 kHz. The site of the experiment was a continental shelf region off the east coast of Korea where the bottom was composed of sandy mud. The mean water depth was 1100 m in the experiment area. Oceanographic data and acoustic data were collected simultaneously during the experiment. It was found that the sound pressure level decreased by 90 dB to 3.4 km and there is little frequency dependence because a strong direct path contributes more than a bottom-reflected path in sound pressure level. At a range between 6 and 7 km, there is a strong bottom-reflected ray path and frequency dependence exists because the bottom reflection loss varies with frequency at a given grazing angle. Sound pressure levels increase as the range increases between 6 and 7 km by 5.4, 1.9, 1.7, and 1.5 dB at frequencies of 1000, 2490, 3990, and 5490 Hz, respectively. (C) 2016 The Japan Society of Applied Physics</P>
-
Jooyoung Hahn,Chunlin Wu,Xue-Cheng Tai 한국산업응용수학회 2010 한국산업응용수학회 학술대회 논문집 Vol.5 No.1
Contemporary TV-Stokes models are based on the so-called two-step methods [1?7]. The first step is to regularize tangential vector field to the level curves of the image and the second step is to reconstruct an image whose gradient fits in the regularized normal vector in the first step. In this paper, we propose the general framework of TV-Stokes models and provide an efficient and fast numerical algorithm based on the augmented Lagrangian method. The proposed model and numerical algorithm have the potential to be used for a number of applications such as image inpainting, image decomposition, surface reconstruction from the sparse gradient, direction denoising, and image denoising. Comparing with the properties of the L²-norm and the L¹-norm in the fidelity term, a proper choice of norms is investigated in each application.We numerically show that the proposed model recovers jump discontinuities of a data and discontinuities of the data gradient while reducing stair-case effect.
-
-
FINE SEGMENTATION USING GEOMETRIC ATTRACTION-DRIVEN FLOW AND EDGE-REGIONS
Jooyoung Hahn,Chang-Ock Lee 한국산업응용수학회 2007 한국산업응용수학회 학술대회 논문집 Vol.2 No.1
A fine segmentation algorithm is proposed for extracting objects in an image, which have both weak boundaries and highly non-convex shapes. The image has simple background colors or simple object colors. Two concepts, geometric attraction-driven flow (GADF) and edge-regions are combined to detect boundaries of objects in a sub-pixel resolution. The main strategy to segment the boundaries is to construct initial curves close to objects by using edge-regions and then to make a curve evolution in GADF. Since the initial curves are close to objects regardless of shapes, highly non-convex shapes are easily detected and dependence on initial curves in boundary-based segmentation algorithms is naturally removed. Weak boundaries are also detected because the orientation of GADF is obtained regardless of the strength of boundaries. For a fine segmentation, we additionally propose a local region competition algorithm to detect perceptible boundaries which are used for the extraction of objects without visual loss of detailed shapes. We have successfully accomplished the fine segmentation of objects from images taken in the studio and aphids from images of soybean leaves.
-
GEOMETRIC ATTRACTION-DRIVEN FLOW FOR IMAGE SEGMENTATION AND BOUNDARY DETECTION
Jooyoung Hahn,Chang-Ock Lee 한국산업응용수학회 2007 한국산업응용수학회 학술대회 논문집 Vol.3 No.2
Noble forces in image segmentation based on active contours models are proposed for capturing objects in the image. Contemplating the common functionality of forces in previous active contours models, we propose the geometric attraction-driven flow (GADF), the binary edge function, and the binary balloon forces to detect objects in difficult cases such as varying illumination and complex shapes. The orientation of GADF is orthogonally aligned with the boundary of object and has the opposite direction across the boundary. It prevents the leakage on the weak edge. To reduce the interference from other forces, we design the binary edge function using the property of orientation in GADF. We also design the binary balloon force based on the four-color theorem. Combining with initial dual level set functions, the proposed model captures holes in objects and multiple junctions from different colors. The result does not depend on positions of initial contours.
연관 검색어 추천
이 검색어로 많이 본 자료
활용도 높은 자료
