二维直线型最小误差阈值分割法

范九伦; 雷博

doi:10.3724/SP.J.1146.2008.01232

二维直线型最小误差阈值分割法

doi: 10.3724/SP.J.1146.2008.01232

范九伦,
雷博

基金项目:

国家自然科学基金(60572133)资助课题

计量
- 文章访问数: 4042
- HTML全文浏览量: 88
- PDF下载量: 1229
- 被引次数: 0
出版历程
- 收稿日期: 2008-10-06
- 修回日期: 2009-03-23
- 刊出日期: 2009-08-19

Two-Dimensional Linear-Type Mnimum Error Threshold Segmentation Method

Fan Jiu-lun,
Lei Bo

摘要

摘要: 一维最小误差阈值法假设了目标和背景的灰度分布服从混合正态分布。考虑到噪声等因素对图像质量的影响，该文在二维灰度直方图上，基于二维混合正态分布假设，给出二维直线型最小误差阈值法的表达式。为了提高算法的运行速度，也给出了快速递推算法。实验表明，二维直线型最小误差阈值法是一个有效的图像分割算法，能够更好地适应目标和背景方差相差较大的含噪图像分割问题。
- 图像处理;阈值分割;最小误差阈值法;二维灰度直方图
Abstract: One-dimensional minimum error thresholding method assumes that the histogram distributions of object and background are governed by a mixture Gaussian distribution. Considering the affects of noise and other factors on image quality, based on the assumption of two-dimensional mixture Gaussian distribution, a two-dimensional linear-type minimum error threshold segmentation method is proposed on two-dimensional gray-level histogram. In order to improve the running speed, a fast recursive formula is also given. Experimental results show that the new method is a valuable image segmentation method which can be well adapted to the cases where the variances of the object and the background are distinctly different and contains noises.

HTML全文

参考文献(1)

Wang S T, Chung F L, and Xiong F S. A novel imagethresholding method based on Parzen window estimate[J].Pattern Recognition.2008, 41(1):117-129[2]范九伦, 赵凤. 基于Sugeno 补的广义模糊熵阈值分割方法[J].电子与信息学报.2008, 30(8):1865-1868浏览[3]Bazi Y, Bruzzone L, and Melgani F. Image tresholding basedon the EM algorithm and the generalized Gaussiandistribution[J].Pattern Recognition.2007, 40(2):619-634[4]Nakib A, Oulhadj H, and Siarry P. Image histogramthresholding based on multiobjective optimization[J]. SignalProcessing, 2007, 87(11): 2516-2534.[5]Otsu N. A thresholding selection method from gray-levelhistograms [J].IEEE Transactions on System Man andCybernetic.1979, 9(1):62-66[6]Kapur N, Sahoo P K, and Wong A K C. A new method forgray-level picture thresholding using the entropy of thehistogram[J].Computer Graphics, Vision and ImageProcessing.1985, 29(3):273-285[7]Kittler J and Illingworth J. Minimum error thresholding[J].Pattern Recognition.1986, 19(1):41-47[8]Mozii F. A note on minimum error thresholding[J]. PatternRecognition Letter, 1991, 12(6): 349-351.[9]Ye Q and Danielsson P. On minimum error thresholding andits implementation[J].Pattern Recognition Letter.1988, 7(4):201-206[10]Fan Jiu-Lun. Notes on Poisson distribution- based minimumerror thresholding[J].Pattern Recognition Letters.1998, 19(5):425-431[11]Abutaleb A S. Automatic thresholding of gray-level picturesusing two-dimensional entropy[J].Computer Graphics,Vision and Image Processing.1989, 47(1):22-32[12]Gong Jian, Li Li-yuan, and Chen Wei-nan. Fast recursivealgorithm for two-dimensional thresholding[J]. PatternRecognition, 1998, 31(3): 295-300.[13]范九伦, 赵凤. 灰度图像的二维Otsu 曲线阈值分割法[J]. 电子学报, 2007, 35(4): 751-755.Fan Jiu-lun and Zhao Feng. Two-dimensional Otsus curvethresholding segmentation method for gray-level image[J].Acta Electronica Sinica, 2007, 35(4): 751-755.[14]李立源, 龚坚, 陈维南. 基于二维灰度直方图的最佳一维投影的图像分割法[J]. 自动化学报, 1996, 22(3): 315-321.Li Li-yuan, Gong Jian, and Chen Wei-nan. The gray-levelimage thresholding method based on the optimalone-dimensional projection of two-dimensional histogram[J].Acta Automa -tica Sinica, 1996, 22(3): 315-321.

施引文献

资源附件(0)

访问统计