• Journal of Korea Society of Industrial Information Systems
• /
• v.3 no.1
• /
• pp.258-268
• /
• 1998

This paper presents two sequential advanced split and merge algorithms and a parallel image segmentation algorithm based on them. First, the two advanced methods are obtained from the combination of edge detection and classic split and merge to solve the inherent problems of the classical method. Besides, the parallel image segmentation algorithm on mesh-connected MIMD system considers three types in the merge stage to reduce the communication overhead between processors, such as intraprocessor merge, interprocessor with boundary merge, and interprocessor without boundary merge. Finally , we prove that the proposed algorithms achieve the improved performance by implementing them.

• Journal of KIISE:Software and Applications
• /
• v.32 no.6
• /
• pp.502-511
• /
• 2005

In this paper, we propose a hybrid analysis method(HAM) based on gray-intensity information from natural scene images. The HAM is composed of GIA(Gray-intensity Information Analysis) and SMA(Split/Merge Analysis). Our experimental results show that the proposed approach is superior to conventional methods both in simple and complex images.

• Journal of the Korea Society of Computer and Information
• /
• v.12 no.2 s.46
• /
• pp.47-52
• /
• 2007

Conventional pixel based region segmentation methods have problems of long processing time and incorrect region split on account of performing region split through comparison of neighboring pixels. In this paper, we propose the method which segments a large size of satellite image effectively using modified centroid linkage method. This method is a sort of region split and merge. The proposed method merges pixels and makes them as a new region through only two directional comparing the current positioning pixel with neighbor ones, if they are satisfied with given conditions. Therefore, this method has less comparing time than the cases of previous ones. The experimental result shows that the proposed method is very efficient because of having less processing time and more exact segmented regions than the previous ones.

• Journal of Korea Multimedia Society
• /
• v.13 no.9
• /
• pp.1314-1321
• /
• 2010

In this paper, we study about the vehicle detection algorithm which is in the process of travelling from the road. An input image is segmented by means of split and merge algorithm. And two largest segmented regions are removed for reducing search region and speed up processing time. In order to detect the back side of the front vehicle considers a vertical/horizontal component, uses an integral image with to apply Haar-like methods which are the possibility of shortening a calculation time, classified with SVM. The simulation result of the method which is proposed appeared highly.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.10
• /
• pp.955-961
• /
• 2010

In this paper, we describe a study of a vehicle detection method based on a Boosting Classifier which uses Histogram of Oriented Gradient (HOG) features and Image Segmentation techniques. An input image is segmented by means of a split and merge algorithm. Then, the two largest segmented regions are removed in order to reduce the search region and speed up processing time. The HOG features are then calculated for each pixel in the search region. In order to detect the vehicle region we used the AdaBoost (adaptive boost) method, which is well known for classifying samples with two classes. To evaluate the performance of the proposed method, 537 training images were used to train and learn the classifier, followed by 500 non-training images to provide the recognition rate. From these experiments we were able to detect the proper image 98.34% of the time for the 500 non-training images. In conclusion, the proposed method can be used for detecting the location of a vehicle in an intelligent vehicle control system.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.04b
• /
• pp.493-495
• /
• 2000

동영상에서 동작물체 영역과 배경 영역을 추출하는 방법에는 크게 원본 영상들의 특징값을 이용하는 방법, 동작벡터 혹은 광류를 이용하는 방법, 그리고 동작벡터와 원본영상을 모두 이용하는 방법의 세가지가 있다. 이중 많이 사용되고 있는 동작벡터를 이용하는 방법에는 히스토그램을 이용하는 방법과 동작벡터의 특징값에 대한 클러스터링을 이용해 분할 하는 방법이 있는데. 이들 기존 방법은 몇가지 문제점을 가지고 있다. 전자는 구현이 간단하나 세부적인 영역분할이 어렵다는 문제점이 있고, 후자는 일반적으로 높은 계산 복잡도를 가지며 초기 클러스터 개수 선정에 문제를 지니고 있다. 본 논문에서는 낮은 계산 복잡도를 가지며 클러스터 할당과 병합된 클러스터 중심 계산에 있어 보다 적응적인 Seeded Split-Merge 클러스터링 방법을 제안한다.

• JSTS:Journal of Semiconductor Technology and Science
• /
• v.12 no.2
• /
• pp.175-185
• /
• 2012

Predictive watershed transform is a popular object segmentation algorithm which achieves a speed-up by identifying image regions that are different from the previous frame and performing object segmentation only for those regions. However, incorrect segmentation is often generated by the predictive watershed transform which uses only local information in merge-split decision on boundary regions. This paper improves the predictive watershed transform to increase the accuracy of segmentation results by using the additional information about the root of boundary regions. Furthermore, the proposed algorithm is processed in a block-based manner such that an image frame is decomposed into blocks and each block is processed independently of the other blocks. The block-based approach makes it easy to implement the algorithm in hardware and also permits an extension for parallel execution. Experimental results show that the proposed watershed transform produces more accurate segmentation results than the predictive watershed transform.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.3E
• /
• pp.118-135
• /
• 2009

The level set based approach is one of active methods for contour extraction in image segmentation. Since Osher and Sethian introduced the level set framework in 1988, the method has made the great impact on image segmentation. However, there are some problems to be solved; such as multi-objects segmentation, noise filtering and much calculation amount. In this paper we address the drawbacks of the previous level set methods and propose an extension of the traditional fast level set to cope with the limitations. We introduce a relationship matrix, a new split-and-merge criterion, a modified Chan-Vese criterion and a novel filtering criterion into the traditional fast level set approach. With the segmentation experiments we evaluate the proposed method and show the promising results of the proposed method.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.24 no.5
• /
• pp.914-922
• /
• 1987

An image coding technique based on a segmentation, which utilizes a simplified description of regions composing an image, is investigated in this paper. The proposed coding technique consists of 3 stages: segmentation, contour coding. In this paper, emphasis was given to texture coding in order to improve a quality of an image. Split-and-merge method was employed for a segmentation. In the texture coding, a linear predictive coding(LPC), along with approximation technique based on a two-dimensional polynomial function was used to encode texture components. Depending on a size of region and a mean square error between an original and a reconstructed image, appropriate texture coding techniques were determined. A computer simulation on natural images indicates that an acceptable image quality at a compression ratio as high as 15-25 could be obtained. In comparison with a discrete cosine transform coding technique, which is the most typical coding technique in the first-generation coding, the proposed scheme leads to a better quality at compression ratio higher than 15-20.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.153-155
• /
• 2003

Laser scanning has become a viable technique for the collection of a large amount of accurate 3D point data densely distributed on the scanned object surface. The inherent 3D nature of the sub-randomly distributed point cloud provides abundant spatial information. To explore valuable spatial information from laser scanned data becomes an active research topic, for instance extracting digital elevation model, building models, and vegetation volumes. The sub-randomly distributed point cloud should be segmented and classified before the extraction of spatial information. This paper investigates some exist segmentation methods, and then proposes an octree-based split-and-merge segmentation method to divide lidar data into clusters belonging to 3D planes. Therefore, the classification of lidar data can be performed based on the derived attributes of extracted 3D planes. The test results of both ground and airborne lidar data show the potential of applying this method to extract spatial features from lidar data.