The Scientific World Journal / 2014 / Article / Tab 7

Research Article

An Efficient Fitness Function in Genetic Algorithm Classifier for Landuse Recognition on Satellite Images

Table 7

Overall accuracy value of each model varying with crossover ways.

Crossover wayIndex
Overall accuracy (%)
DBIFCMIPASIDBFCMIAverage

P174.3 71.6 72.9 74.6 73.3
P271.6 73.2 64.9 75.5 71.3
P370.8 72.5 71.5 72.5 71.8
P468.8 74.5 68.9 74.7 71.7
P570.5 70.2 70.9 75.0 71.6
P668.7 64.2 56.0 75.0 73.3
Standard deviation1.23.01.40.2

String length: 8, population: 30, selection way: roulette wheel selection, crossover rate: 0.8, and mutation rate: 0.003.