• KIPS Transactions on Software and Data Engineering
• /
• v.3 no.5
• /
• pp.195-200
• /
• 2014

Typical methods of active contour model for object contour extraction are snake and level. Snake is usually faster than level set, but has limitation to compute topology of objects. Level set on the other hand is slower but good at it. In this paper, a new object contour extraction algorithm to use advantage of each is proposed. The algorithm is composed of two main steps. In the first step, snake is used to extract the rough contour and then in the second step, level set is applied to extract the complex contour exactly. 5 binary images and 2 natural images with different contours are simulated by a proposed algorithm. It is shown that speed is reduced and contour is better extracted.

• The KIPS Transactions:PartB
• /
• v.17B no.5
• /
• pp.375-380
• /
• 2010

Most of previous algorithms of object boundary extraction have been studied for extracting the boundary of single object. However, multiple objects are much common in the real image. The proposed algorithm of extracting the boundary of each of multiple objects has two steps. In the first step, we propose the fast method using the outer and inner products; the initial contour including multiple objects is split and connected and each of new contours includes only one object. In the second step, an improved active contour model is studied to extract the boundary of each object included each of contours. Experimental results with various test images have shown that our algorithm produces much better results than the previous algorithms.

• KIPS Transactions on Software and Data Engineering
• /
• v.4 no.8
• /
• pp.309-314
• /
• 2015

One of typical methods for extracting facial features from face images may be snake. Although snake is simple and fast, performance is very much affected by the initial contour and the shape of object to be extracted. In this paper, the enhanced snake is proposed to extract better facial features from 6 lip and mouth images as snake point is added to the midpoint of snake segment. It is shown that RSD of the proposed method is about 2.8% to 5.8% less than that of Greedy snake about 6 test face images. Since lesser RSD is especially obtained for contours with highly concavity, the contour is more accurately extracted.

• The Transactions of the Korea Information Processing Society
• /
• v.5 no.11
• /
• pp.3010-3023
• /
• 1998

GA has a fast convergence speed in searching the one point around optimal value. But it's convergence time increase in searching the region around optimal value because it has no regional searching mechanism. BP has the tendency to converge the local minimum because it has global searching mechanism. To overcome these problems, a method in which a genetic algorithm and a back propagation are applied in turn is proposed in this paper. By using a genetic algorithm, we compute optimal synaptic strength and offset value. And then, these values are fed to the input of the back propagation. This proposed method is superior to each above method in improving the convergence speed.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.04b
• /
• pp.19-21
• /
• 2001

본 논문은 객체의 유사성 비교를 위해 객체의 모양을 표현하는 한 가지 특징인 윤곽선상의 우세 점들을 찾는 재귀적 윤곽선 근사 알고리즘을 제안한다. 이 알고리즘은 같은 모양의 개체에 대하여 그 객체의 무게 중심을 이용하여 항상 일정한 특정 시작점을 찾음으로써 동일한 우세 점들을 재귀적으로 빠른 수행 시간에 찾는다. 또한 이 알고리즘은 열린 곡선, 닫힌 곡선 및 다각형 등 어떤 모양의 평면 도형에도 모두 적용 가능하다. 제안 알고리즘의 평균 시간 복잡도는 O(nlogn)이다.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2002.05d
• /
• pp.1087-1092
• /
• 2002

본 논문에서는 결정 트리 학습을 이용하여 뇌 MRI 시리즈를 분류하는 시스템을 제안한다. 영상으로부터 얻을 수 있는 정보에는 두 종류가 있다. 하나는 크기, 색상, 질감, 윤곽선 등 원 영상으로부터 직접 얻을 수 있는 하위레벨(low-level) 특징들이고, 다른 하나는 특정 개체의 존재유무, 여러 부위 사이의 공간적 관계 등 분할된 영상들에 대한 해석을 통하여만 얻을 수 있는 상위레벨(high-level) 특징들이다. 영상을 의미에 따라 분류하기 위해서는 학습 및 분류가 상위레벨 특징들을 기반으로 수행되어야 한다. 제안된 시스템에서는 결정 트리 학습을 이용하여 영상을 구성하는 요소를 학습하고 분류하며 그에 따라 영상 시리즈를 대표할 수 있는 상위레벨 특징을 추출하였다. 정상, 뇌경색, 뇌종양이 있는 뇌 MRI 시리즈에 대하여 분류 실험을 수행하였으며, 그 결과를 설명 하였다.

• Journal of IKEEE
• /
• v.16 no.2
• /
• pp.95-101
• /
• 2012

A novel approach for the reconstruction of 3D building model from aerial image data is proposed in this paper. In this approach, a Centroid Neural Network (CNN) with a metric of line segments is proposed for connecting low-level linear structures. After the straight lines are extracted from an edge image using the CNN, rectangular boundaries are then found by using an edge-based grouping approach. In order to avoid producing unrealistic building models from grouping lined segments, a hierarchical grouping method is proposed in this paper. The proposed hierarchical grouping method is evaluated with a set of aerial image data in the experiment. The results show that the proposed method can be successfully applied for the reconstruction of 3D building model from satellite images.