Abstract
This paper proposes a new method to estimate the distribution of farm product fields by analyzing the satellite image. In this study the konjac field is dealt with mainly as the farm product field. However, it is expected that the result of this study is applicable to the other farm products as well as the konjac. The procedure of the land cover classification by the proposed method is as follows.
1) The likelihood values are estimated by using the pixel values of the band 1-3 of the satellite image, the NDVI and the NDCI as the evaluation index.
2) The land cover classification is performed by judging the likelihood values with the standard likelihood value corresponding to the reliability.
The field survey and satellite image photography were executed simultaneously in the region where the farm product fields crowd. The proposed method and the maximum likelihood method were applied to the obtained satellite image, and the classification analysis of the satellite image was executed. It was confirmed that the both method results correspond with the field survey result with a fair degree of precision. It became clear that the proposed method gives more satisfactory result than the maximum likelihood method by comparing the both methods.
