Abstract
On-board real-time anomaly detection has always been a challenging task in hyperspectral imaging analysis as it requires low computational complexity. Most of the existing anomaly detection algorithms inevitably trade off intensive computational complexity for high detection accuracy. This article presents a fast spectral-spatial anomaly detection algorithm with low complexity in hyperspectral images (HSIs) using morphological reconstruction and a simplified guided filter (Fast-MGD). Since the simple filtering techniques are applied, it is therefore feasible to achieve a field programmable gate array (FPGA)-based hardware implementation. More precisely, an effective deeply pipelined acceleration scheme is developed adopting high-level synthesis to support HSIs that are acquired over different scenes with different sizes and spectral bands. Experimental results show strong advantages of the proposed FPGA-based Fast-MGD in processing speed and resource consumption, while a high detection accuracy is remained. Its applicability in on-board real-time processing is demonstrated and verifie.
Highlights
O WING to the richness and abundance in spectral–spatial information, hyperspectral images (HSIs) acquired by hyperspectral imaging have been widely used in various applications including classification [1], target or anomaly detection [2], etc
We propose a fast spectral–spatial anomaly detection algorithm based on morphological reconstruction and the simplified guided filtering (Fast-MGD)
The pixel derived from the Data Loader and Distributor is first stored in FIFO performing with 512 bit width, and the high-bit data are transferred to the Q Spectral Integration units, thereby reducing processing time through full use of the bandwidth of DDR3 SDRAM
Summary
O WING to the richness and abundance in spectral–spatial information, hyperspectral images (HSIs) acquired by hyperspectral imaging have been widely used in various applications including classification [1], target or anomaly detection [2], etc. Benefiting from the fact that anomaly detection neither requires any prior information nor relies on the complex preprocessing like atmospheric and radiometric correction, hyperspectral anomaly detection shows its high applicability for real-time processing on satellites. Aiming at solving problems of on-board real-time processing of hyperspectral anomaly detection, the motivation of our work is to develop an algorithm and hardware structure for both high accuracy and low complexity. We propose a fast spectral–spatial anomaly detection algorithm based on morphological reconstruction and the simplified guided filtering (Fast-MGD).
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
