30-06-2012, 01:37 PM
Research on High-Performance Remote Sensing Image Real-Time Processing System
[attachment=26282]
Abstract
Research and discuss the target detection algorithms
and flows in remote sensing image. Different from other
simulation detection algorithms, we added span-tile
determination steps in an actual implementation. Based on the
analysis of remote sensing image real-time processing system
features, design and implement a high-performance real-time
target detection signal processing system. The system consists of
an amount of dedicated floating point DSPs formed a processing
element (PE) and high-speed serial switching architecture which
established the communication links between PEs.
INTRODUCTION
In recent years, focusing on massive data storing and
processing, high-speed data flows, multisource, the remote
sensing image information processing technologies developed
rapidly [ 1]. Traditional PCs never satisfy the real-time
processing demanding. In order to largely increase the
computing speed, processing performance and execution
efficiency, multi-core technologies and parallel computing
theories were adopted [2]. Nowadays, there are three types of
mainstream processing systems used in remote sensing image
processing, which described as follows.
DETECTION ALGORITHM
The overall framework of the ship detection algorithm is
summarized in Fig.4, and the following describe in detail
about the major stages of the proposed algorithm. Remote
sensing images are passed through the preprocessing,
pre screening and discrimination.
SYSTEM ANALYZE AND DESIGN
According to the functional requirements, remote sensing
image real-time processing system can be divided into three
main subsystems, as illustrated in Fig. 5. They are satellite
data formats rescheduling system, image processing system
and comprehensive processing system.
CONCLUSION
The main objective of this paper was to propose a highperformance
platform for remote sensing image real-time
processing system, satisfy the demanding of target detection
algorithms and the corresponding work flows. Utilized highperformance
parallel DSP technologies and high-speed serial
switching architecture, the performance and flexibility of the
system increased largely. Another objective was to propose
and add the span-tile target detection steps, which are different
from traditional none consideration algorithms.
[attachment=26282]
Abstract
Research and discuss the target detection algorithms
and flows in remote sensing image. Different from other
simulation detection algorithms, we added span-tile
determination steps in an actual implementation. Based on the
analysis of remote sensing image real-time processing system
features, design and implement a high-performance real-time
target detection signal processing system. The system consists of
an amount of dedicated floating point DSPs formed a processing
element (PE) and high-speed serial switching architecture which
established the communication links between PEs.
INTRODUCTION
In recent years, focusing on massive data storing and
processing, high-speed data flows, multisource, the remote
sensing image information processing technologies developed
rapidly [ 1]. Traditional PCs never satisfy the real-time
processing demanding. In order to largely increase the
computing speed, processing performance and execution
efficiency, multi-core technologies and parallel computing
theories were adopted [2]. Nowadays, there are three types of
mainstream processing systems used in remote sensing image
processing, which described as follows.
DETECTION ALGORITHM
The overall framework of the ship detection algorithm is
summarized in Fig.4, and the following describe in detail
about the major stages of the proposed algorithm. Remote
sensing images are passed through the preprocessing,
pre screening and discrimination.
SYSTEM ANALYZE AND DESIGN
According to the functional requirements, remote sensing
image real-time processing system can be divided into three
main subsystems, as illustrated in Fig. 5. They are satellite
data formats rescheduling system, image processing system
and comprehensive processing system.
CONCLUSION
The main objective of this paper was to propose a highperformance
platform for remote sensing image real-time
processing system, satisfy the demanding of target detection
algorithms and the corresponding work flows. Utilized highperformance
parallel DSP technologies and high-speed serial
switching architecture, the performance and flexibility of the
system increased largely. Another objective was to propose
and add the span-tile target detection steps, which are different
from traditional none consideration algorithms.