As part of the fundamental studies of the multiple image processing, we have been developing the NOAA AVHRR three-dimensional analysis system, in which we encountered the problem of drawing the three-dimensional histogram on the two-dimensional CRT. In this case there are two factors: One is the coordinate position given by the brightness level of the three image channels of NOAA AVHRR, the other is the frequency of the points within the images. In order to decide the semi-optimum value to display the processed three-dimensional histogram, we tried to relate between frequency and the total number of pixels. We found that standard deviation of them is reasonable. In the results of analysis using this method, the difference of major components of images are displayed in the three-dimensional histogram. And thus we were able to identify the features of the histogram corresponding to the physical entities like, sea, land, cloud, etc. by a process known as clustering.
|ジャーナル||IEEE Transactions on Geoscience and Remote Sensing|
|出版物ステータス||Published - 1991 9|
ASJC Scopus subject areas
- Electrical and Electronic Engineering
- Earth and Planetary Sciences(all)