For the traditional method of hyperplane segmentation

the location of hyperplane in data space was determined by statistical method. In the case of the statistical value of regions is smaller than thoes within the region

the statistical method was not effective. The character of genetic algorithm is global searching optimally. Taken this mathematical advantage

the location of Hyperplane could be located easily. In the paper

we use EOS/MODIS imagery data as an example to introduce this method in detail. We demonstrate that genetic algorithm can be used to produce segmentation result of remote sensing data. Using the parameters of the hyperplanes encoded in the chromosome

the region in which each training pattern point lies is determined by hyperplanes equation. Then the fitness is decided. After computing the fitness

the genetic operators of selection

crossover and mutation are applied to generate a new population of chromosomes. Then

after some process of this

the fitful hyperplane will be generated by this process. At last

the image was divided by the most fitful hyperplane equations. After the introduction of this method and its application

the comparative results with Maximum likelihood（ML）method in ERDAS IMAGINE 8.4 software package are given in section 4.As the paper shows

the genetic method is clearly better than ML method. It need to note that the realization is based on the Windows and Turbo c 2.0. We would have better results in VC++.based on Windows XP.