文档下载 文档下载导航
设为首页 | 加入收藏
搜索 请输入内容:  
 导航当前位置: 文档下载 > 所有分类 > TWO-DIMENSIONAL ENTROPY METHOD BASED ON GENETIC ALGORITHM
侵权投诉

TWO-DIMENSIONAL ENTROPY METHOD BASED ON GENETIC ALGORITHM

Abstract-Image segmentation represents the first step in many applications in image understanding. Thresholding is a simple and important in image segmentation. One-dimensional entropy does not take into account the spatial correlation between the pixels i

ψ(s,t)=H(O)+H(B)=lnPst+HstP+ln(1 Pst)+(HL 1L 1 Hst1 P)

stst

=lnPst(1 Pst)+

Hst

st

+

(HL 1L 1 Hst

1 Pst).

The algorithm then searches for the values of

(s,t) that maximizes ψ(s,t), viz.,

s,t∈G

t =ArgMaxψ(s,t).

3. Genetic Algorithm

Genetic Algorithm is created by analogy with the processes in the reproduction of biological organisms[7]. It can be classified as guided random search evolution algorithms that uses probability to guide its search.

Simple Genetic Algorithm (SGA) employs the following genetic operators in a search procedure: proportional reproduction, one-point crossover, and simple mutation. It maintains a constant-sized “population” of candidate solutions, known as individual. The initial seed population from which the genetic process begins can be chosen randomly or on the basis of heuristics, if available for a given application. At each iteration, known as a “generation”, each individual is evaluated and recombined with others on the basis of its overall quality of “fitness”. In this paper, we improve the proportional reproduction by applying Remainder Stochastic Sampling with Replacement (RSSR). Two-point crossover is employed to replace one-point crossover. The elitist strategy is also employed to assure the Genetic Algorithm’s convergence. 3.1 Improvement to basic reproduction operator

The proportional reproduction operator in SGA decides the expected number of times an individual is selected for recombination, which is proportional to its fitness relative to the rest of the population.

Let

N

be the population size,

fiti be the ith individual’s fitness, the possibility for this ith individual is

defined as

pi=

fiti

∑fit

i=1

N

(i=1,2, ,N).

i

An obvious drawback of “proportional reproduction” is that individuals with high fitness may not be selected and reproduced especially when the generation turns to be relatively small. The selection error is caused by the random selection techniques.

Remainder Stochastic Sampling with Replacement (RSSR) [9] is employed in the paper to improve the basic proportional reproduction operator. RSSR reproduction operates as follow steps:

1) Let

Ni be the expected number of times an individual is selected for recombination, which may possibly

not be an integer.

Ni=

N fiti

∑fit

i=1

N

(i=1,2, ,N).

i

2) Let

Ni be the integer part of Ni so as to construct ∑ Ni individuals of the next offspring.

i=1

N

上一页第4页

你可能喜欢

热门文档

相关文档

站点地图 | 文档上传 | 侵权投诉 | 手机版
本站所有资源均来自互联网,本站只负责收集和整理,均不承担任何法律责任,如有侵权等其它行为请联系我们.
文档下载 Copyright 2013 doc.guandang.net All Rights Reserved.  -  email