It is an important part of remote sensing application to investigate and monitor the natural disasters using remote sensing technology. Based on the plenty of data and author′s working experience abroad, the main works and their benefits applying remote sensing technology to monitor , forecast disaster and to reduce the loss of disaster in developed countries and some developing countries were introduced briefly in this paper.

Remote Sensing (RS), Global Positioning System (GPS), and Geographical Information System (GIS) are being rapidly developped, in the meantime, they are forging ahead toward the Integration and resulting in the so-called “3S-Integration”. The 3S-Integration gives a new and great potential technical support to the investigation and research of seismic hazards. In this paper, the application of 3S-Techniques to investigate and study seismic hazards are briefly described, and , in the point view of Integration emphatically, the use model and realizations of 3S-Integration in the futural investigation and research of seismic hazards aer discussed.

This paper deals with how to use the color infrared aerial photos and TM data to extract the major factors which effected soil erosion, how to use method of levels analysis to make the expert right weight model and to ascertain the grades of soil erosion. It also analyzes the feature of soil erosion and puts forward some harness measurements in Panzhihua area.

Made use of remote sensing information which was greatly integrated and highly summarized and combined with relative hydrogeological data in Enkou coal mine, the main factors affecting karstification of land surface were analyzed. The image features of lithological character, geological structure, hydrogeological conditions and karst landform were discribed. Based on what mentioned above, this paper put forward the interpretation keys of karstification and pointed out that this qualitative interpretation can be the base of the quantitative study of karstification level.

Using one year NOAA-AVHRR data of channel 1, 2, 4 and the method of image enhancement, we study the temprature, area, algal and silt of Dianchi Lake. Though the resolution of NOAA satellite is not very high, it has superiority in monitoring dynamic change of lake water.

Based on the interpretation of satellite and aerial images and the analysis of field surveying data about the channel evolution and bankline resources, the distribution, use status and change tendency of the bankline resources in Wuhu port of the Yangtze river have been found out in this paper. For the purpose of providing scientific basis for proper using the bankline resources, this paper has also made a assessment to the bankline resources use.

Supported by GIS, the author of this paper established a spatial database using TM images of two periods and some forest surveying maps, and acquired a shift matrix about the change of forest resources in Liupanshan mountain area, then predicted its evolution tendency using the Markov-chain model.

It is difficult to geometrically rectify remote sensing images in personal computer because of the large data. A PC-based fast algorithm of geometric rectification is presented to solve this problem. Three techniques are adopted in the algorithm: (1) fast calculating the polynormal by separating column items from rows; (2) accessing optimal rectangle sub-image according to the allocated computer memory; (3) processing all bands simultaneously. These techniques make the algorithm fast and suitable for images with any size. The implementation and performance of the algorithm are also discribed in this paper.

A method of semilinear feedforward net applied to the classification of remote sensing image was discussed in this paper, Satisfactory results were obtained in extracting the planted rice area in Huazhuang and Yangjian towns of Wuxi county from TM data of July 23, 1991 by this method.

This paper discusses the methods of automatic extraction of digital elevation model (DEM) from remote sensing image. A system of scheme is established to extract DEM from SPOT image. The scheme improves the multi-level image matching techniques. Several fast algorithms suitable for personal computer are adopted. A software system is developed in PC acording to the scheme. The results of practical tests indicate that the scheme is feasible and has high accuracy and reliability. The software system is practicable and runs fast.