Color image segmentation based on region growing algorithm

被引：0
作者：
机构：
[1] Graduate School of Computer Science and Engineering, The University of Aizu, Fukushima
来源：
Shin, J. (jpshin@u-aizu.ac.jp) | 1600年 / Advanced Institute of Convergence Information Technology卷 / 07期
关键词：
Color Image; RGB; Segmentation;
D O I：
10.4156/jcit.vol7.issue16.18
中图分类号：
学科分类号：
摘要：
In this paper, color image segmentation that recognizes objects in an image with region growing algorithm is discussed to be applied to moving object recognition algorithm. The algorithm that dynamically calculates the number of image segments by region growing algorithm is obtained. Initial segments start at one pixel and then connect to others by a conventional algorithm. The dividing image algorithm uses the area variance to obtain initial segments and to improve process effectiveness. The best color model for growing segments is discussed through these experiments. The RGB color model is the best color model for dividing segments and the L*a*b* color model is the best for connecting segments. As a result of image segmentation using region growing algorithm, wide area segments are recognized without shadows, JPEG noise, and mosquito noise. The shadowing segments are also recognized. The proposed algorithm is applied to edge processing and background subtraction to define the proposed algorithm features. As a result, objects are recognized with less noise by using the proposed algorithm.
引用
收藏
页码：152 / 160
页数：8
相关论文
共 12 条
  • [1] Hamamoto T., Hangai S., Miyauchi K., A Study on Region Growing Method for Shape Acquisition, 29, pp. 189-190, (1993)
  • [2] Hiroshi S., Color Image Segmentation by Clustering Technique in RGB space, 38, pp. 31-34, (2004)
  • [3] Shiji A., Hamada N., Color Image Segmentation Method Using Watershed Algorithm and Contour Information, The transactions of the Institute of Electronics, Information and Communication Engineers. D-II, J83-D-II, 2, pp. 593-600, (2000)
  • [4] Takahashi K., Abe K., Color Image Segmentation Using ISODATA Clustering Algorithm, Institute of Electronics, Information, and Communication Engineers, J82-D-II, 4, pp. 751-762, (1999)
  • [5] Takeharu I., Tetsuya Y., Junji M., Color Quantization Based on 3D Histogram and Segmentation of Color Images, ITE technical report, 28, 7, pp. 101-106, (2004)
  • [6] Tamaki T., Yamamura T., Ohnishi N., Object-Region extraction in a image, Technical report of IEICE PRMU, 97, 501, pp. 63-70, (1998)
  • [7] Sudo O., Tsuruma K., Maeda J., Segmentation of Color Images Using Texture Features and Fuzzy Region Growing Algorithm, Institute of Electronics, Information, and Communication Engineers, 103, 640, pp. 89-94, (2004)
  • [8] Yamamoto K., Osaki H., Maeda J., Rough Segmentation of Color Images Using Texture Features, Institute of Electronics, Information, and Communication Engineers, 28, 7, pp. 95-100, (2004)
  • [9] Yoshida T., Background Differencing Technique based on the Status of Reference Pixels, Institute of Electronics, Information, and Communication Engineers, J88-A, 11, pp. 1226-1234, (2005)
  • [10] Watanabe T., Sasaki T., Kimura A., An Improvement of the Optimal Filter ISEF for Edge Detection, Institute of Electronics, Information, and Communication Engineers, J87-D-II, 3, pp. 914-918, (2004)
12