Multi-FPGA (Field Programmable Gate Array) cooperative on-satellite real-time lightning image processing system and method

Abstract

The invention provides a multi-FPGA (Field Programmable Gate Array) cooperative on-satellite real-time lightning image processing system and method. The multi-FPGA cooperative on-satellite real-time lightning image processing system comprises a real-time lightning detecting-processing module, a background image data processing module and a lightning signal and background image encoding module, andcan process original image data being 500 frames per second and up to 10.9Gbps in data rate in real time to finish lightning detection processing. A coordinative parallel image processing technologyimplemented by eight high-performance SRAM type FPGAs and four anti-fuse FPGAs is adopted, so that a lightning signal can be detected and identified effectively. Meanwhile, original background data blocks are extracted flexibly in real time to implement the function of combined encoding of lightning signal data and the background data blocks. The original image blocks are extracted and transmitteddownwards to the ground through the anti-fuse FPGAs, and the recognition and extraction of a transient target are finished by the SRAM type FPGAs. Through adoption of the multi-FPGA cooperative on-satellite real-time lightning image processing system, the on-orbit target recognition efficiency is increased. The multi-FPGA cooperative on-satellite real-time lightning image processing system can meet multipurpose application, has high radiation resistance, and is suitable for being applied to high-altitude satellites.

Description

technical field [0001] The invention belongs to the field of on-board real-time image processing, and proposes a multi-FPGA cooperative on-board real-time lightning image processing system and method, which can be used for real-time processing of on-orbit camera image data to detect and identify transient lightning signals. Background technique [0002] Lightning detection plays an important role in meteorological observation. By monitoring the location, intensity and frequency of lightning activity, we can deepen the understanding of the distribution, structure and changes of strong convective weather on land and sea, and timely and accurately forecast disasters such as heavy rain and hurricanes. Natural weather can greatly reduce the loss of human life and property on a large scale. Lightning observation is to obtain the original image through a 500 frames per second high-speed imager. The amount of data is large. The massive data is limited by the bandwidth of the satelli...

AI Technical Summary

Problems solved by technology

Lightning observation is to obtain the original image through a 500 frames per second high-speed imager. The amount of data is large. The massive data is limited by the bandwidth of the satellite-ground link and cannot be transmitted to the ground receiving station; at the same time, the limited image data is transmitted to the ground for reprocessing. The timely detection, identification and processing of such transient targets must rely on on-orbit processing to detect and identify lightning signals

For special image processing such as lightning detection, conventional on-board image processing methods cannot solve such problems. In addition, space-borne equipment has strict constraints on volume and power consumption, and ground-based large-scale computer processing methods are not applicable.

Images