The paper first studied the geometric correction of historical CORONA satellite imagery in the 1960s and used the historical imagery to extract KRD (karst rocky desertification). The study area is located in the karst region of Guangxi Province, China. Finally, we used the Landsat-5 imagery to extract rocky desertification in 2005, then we could find the changes of the karst rocky desertification in Guangxi from 1960s to 2005 about nearly 40 years. And comparison analysis was conducted and the results showed that, over the 40 years, Guangxi karst rocky desertification area has significantly changed. Guangxi has typical karst environment and is one of the most serious areas of rocky desertification in southwest China provinces, thus our research on this has great practical significance.
A new generation of flash LiDAR sensor called <i>GLidar-I</i> is presented in this paper. The GLidar-I has been being developed
by Guilin University of Technology in cooperating with the Guilin Institute of Optical Communications. The GLidar-I
consists of control and process system, transmitting system and receiving system. Each of components has been designed
and implemented. The test, experiments and validation for each component have been conducted. The experimental results
demonstrate that the researched and developed GLiDAR-I can effectively measure the distance about 13 m at the accuracy
level about 11cm in lab.
This research dissertation presents a new decision tree induction method, called co-location-based decision tree (CL-DT),
to extract exposed carbonate rocks in karst rocky desertification area. The proposed algorithm utilizes co-location
characteristics of multiple feature parameters, including landmarks spectral attribute, vegetation fraction, land surface
temperature and soil moisture content, etc., and spatial attributes of various landmarks in desertification area. This paper
first presented multiple feature parameters co-location mining algorithm, including attributes data selection,
determination of rough candidate co-locations, determination of co-locations, pruning non-prevalent co-locations, and
inducing co-location rules, and then focused on developing the algorithm of co-location decision tree, which including
non-spatial attributes data selection, multiple feature parameters co-location modeling, node merging criteria, and colocation
decision tree induction. The paper uses Landsat-5 TM images covering the whole Du‟an city in China as the data
to verify the proposed method. The experimental results demonstrated that (1) Compared to traditional decision tree, the
proposed multiattribute co-location decision tree has higher accuracy and can make better decision; (2) The training data
can be fully played roles in contribution to decision tree induction.