Shamoh area is located east-north part of Hokkaido prefecture and suger beet field widely spread between Shari town and Abasiri city. Suger beet was damaged by dry weather in 1984, and was not so damaged in 1988. In this paper, we analyze the suger beet yields and make yields map in 1984 and 1988 using satellite data, Landsat MSS and TM. Next, we analyze soil water contents and soil humid in the field using Landsat TM. As a result, Typic Brown Lowland soils can estimate soil water contents, but Stratic Regosolic Andosols cannot. Soil humid can be estimated and we make distribution map. Using suger beet yields map, soil humid map and soil map, we analyze the relation of soil type and yields. As a result, soil types which are damaged heavily by dry weather compared usual year are distinguished.
Maximum likelihood classifier that is often used in classification of satellite images assumes the distribution of each class to Gaussian. Such linear classifier can classify correctly when the case that classification probability of each class is exclusive. Remotely sensed data, however, belong to several classes and have non-linear separable condition. To improve the classification accuracy of non-linear separable data, the application of the single-step multi-layer back propagation neural networks have been studied by many researchers. In this paper, multi-step multi-layer neural networks, so called cooperative learning neural networks, are proposed to classify the non-linear separable satellite data. The cooperative learning neural network consists of extraction networks for each class and an unification network which unifies the extracted values. The unification network is also used for unification of different environments such as time-series data or neighboring regions. The result of the classification of LANDSAT TM data of Nagoya city using the cooperative learning neural network is introduced. Classified image is compared with the detailed digital land cover information (TDT-112) and the images classified using single-step multi-layer neural network, maximum likelihood classifier and fuzzy set reasoning. As the result of the comparison, the cooperative learning neural network classify the remote sensing data more exactly than the other methods.
Using the NOAA-HRPT data received at the Tohoku University, a satellite image database for regional researches is produced at the Computer Center of the university. The database is called“Tohoku Image Database: TIDAS”. The TIDAS is open for public, and can be accessed through computer networks. The albedo and brightness temperature processed from the AVHRR channel 2 and channel 4 are geocoded, and sent to the Computer Center for TIDAS production every day. The TIDAS contains the satellite images in 99.2% of the days from April, 1990 to the present. This paper describes the overall features of the TIDAStogether with the related background.