Curvelet Transform Based Hyperspectral Image Compression with Listless Set Partitioned Compression Algorithm for Unmanned Aerial Vehicle Image Sensor

Over the past several years, we have witnessed remarkable progress in hyperspectral (HS) images taken by unmanned aerial vehicles. The HS image is very high in spectral resolution in many narrow contiguous bands. The HS image compression becomes necessary to effectively handle large amounts of remote sensing data for storage and communication purposes. In recent years, many compression algorithms have been proposed to achieve a high compression ratio, but they either suffer from coding efficiency or coding memory or coding complexity. Transform-based Hyperspectral Image Compression Algorithm (HSICA) exploits the correlation that exists between the pixels or frames of the HS image and works with both types of compression. The wavelet transform-based set partitioned HSICAs are a special type of transform based HSICAs that use data-dependent link lists or image size-dependent state tables to track the significance of sets or coefficients and have better compression performance than other HSICAs due to the exploitation of the HS image redundancies. The proposed compression algorithm 3D-Low Memory Zerotree Coding (3D-LMZC) uses the curvelet transform to improve directional elements and better the ability to represent edges and other singularities along curves. The objective of the proposed HSICA is to achieve a high compression ratio while simultaneously representing HS images at a variety of scales and directions. This will allow for the provision of compressed HS images of a high quality. The results of the experiments reveal that the suggested approach has a low coding memory demand, and compared to other state-of-the-art compression algorithms, it achieves an increase in coding gain of approximately 5%.