Specialized hyper spectral image compression using 3 dimensional factorization technique and its thriving performance analysis
S Hariharasudhan, Dr. B Raghu
Data compression is a fundamental and original requirement of digital information sequence of storage, broadcast and process of recovery. A digital image figures a momentous part in multi-media systems. This research work is highlighted on the expansion of a non-iterative with positive cum matrix factorization technique for image represented data compression. Though positive matrix factorization is a well-structured and recognized technique for image data compression, this technique is noteworthy on iterative method for factorization. In this research job, a very useful non-iterative technique for positive matrix factorization has been urbanized using Asymmetric Haar Discrete Wavelet transformation and experienced and tested with medical images. This 2-Dimensional matrix factorization system has also been stretched to 3D matrix factorization and useful for hyper-spectral image compression. Consequently a disparity signaling based BPSK (Bipolar Phase Shift Keying) system has been introduced for the transmission of factorized coefficients and tested for its performance in preservative White Gaussian Noise channel causing Rayleigh channels. A comprehensive statistical analysis has been prepared on the presentation of differential signaling related BPSK in communication channels exaggerated by Gaussian noise. The highly developed communication system has also been experienced using multi medical image data. The differential signaling related BPSK communication has a natural ability for partial discovery of errors without any transparency bits. With suitable and appropriate channel standard coding schemes at the value of overhead bits, the proposed system will be executed in improved manner.