Work

GPU Implementation of Three-Stage Prediction with Adaptive Search Threshold for Hyperspectral Image Compression

Year: 2017
Type: article
Abstract: For the huge size of hyperspectral images with hundreds of bands, compression is necessary to save storage space and transmission time.Three-stage prediction with adaptive search threshold (TSP-AST), which consists of third-order interband predictor, bi-directional pixel search, backward pixel search with adaptive search threshold, and entropy coding, has been proven to be an effective lossless compression method for hyperspectral images, but its computational complexity is very high which makes its application to time-critical scenarios quite limited.In order to improve the computational efficiency of the algorithm, a parallel implementation of the TSP-AST algorithm is presented using compute unified device architecture (CUDA) on graphics processing units (GPUs) by exploiting the data parallel characteristics of the three prediction stages.The GPU parallel implementation is compared with the serial and multicore implementations on CPUs.Experimental results based on real hyperspectral images reveal remarkable acceleration factors and real-time computing performance in our four-GPU-based implementation of the TSP-AST compression scheme, while retaining exactly the same compression performance with regard to the serial and multicore versions of the compressor.
Language: English
Open Access status: bronze
OpenAlex ID: W2931743538
FWCI:
Cites: 23
Cited by:
Related to: 10