A Study of Improvements in the Efficiency of High-Resolution Remotely Sensed Image Parallel Processing
This paper presents the information extraction method based on feature-unit of high-resolution remotely sensed image. In order to extract all kinds of information from remotely sensed image efficiently, this paper presents the research idea of image coarse-classification based on large scale and precise-segmentation based on small scale; in order to improve the speed of image processing, we take parallel computing method to solve this problem. For the data partition method of remotely sensed image parallel processing, this paper summarizes the general data partition methods and presents a new data asymmetric partition method; for the data transfer method of parallel computing based on image database, this paper gives a new data transfer method to avoid the pleonastic data transfer. At last this paper gives the implementation of the research idea based on MPI and analyzes the efficiency, and the results show that the new methods can improve the system efficiency efficiently.
parallel processing information extraction feature-unit data partition remotely sensed image database
Zhanfeng Shen Jiancheng Luo Dongping Ming
Institute of Remote Sensing Applications, Chinese Academy of Sciences, Beijing 100101, P.R.China
国际会议
北京国际地理信息系统学术讨论会第七届会议（7th International Workshop Geographical Information System
北京
英文
407-416
2007-09-14（万方平台首次上网日期，不代表论文的发表时间）