Strip Based Embedded Coding of Wavelet Coefficients for.pdf

Strip Based Embedded Coding of Wavelet Coefficients for Large Images Raghunadh K Bhattar EE Department Indian Institute of Science Bangalore raghu@sac.ernet.in K.R.Ramakrishnan EE Department Indian Institute of Science Bangalore krr@ee.iisc.ernet.in K.S.Dasgupta ADCTG Space Applications Centre Ahmedabad. ksd@sac.ernet.in Abstract Wavelet based embedded coders such as EZW and SPIHT require full wavelet transform of the image to be buffered for coding. Further, since the transform coefficients are required to be stored in high precision, the buffering requirements are prohibitively high for large images, such as remote sensing images, space borne images, medical images etc. In this paper we investigate, embedded coding of wavelet coefficients using zero trees, with reduced memory requirements, for large images. The wavelet coefficients are buffered in a ‘strip buffer’, capable of holding few lines of wavelet coefficients from all the subbands belonging to the same spatial location. For above applications, we also develop a pipeline architecture for real time embedded coding. 1 Introduction Images acquired through remote sensing satellites are generally large in size. These images, acquired line-by- line by an optical sensor, are to be transmitted in compressed form on a constant bit rate channel to ground station for archival and analysis purpose. Wavelet based compression techniques, such as EZW and SPIHT have become benchmark techniques for state of art compression. Besides, better compression and better image quality, they provide embedded bitstream, which can be truncated at any point and reconstruction carried out at varying quality. The price to be paid for these advantages is in the form of large memory requirements. For coding, the transform coefficients of full image need to be buffered in high precision, thus making memory requirements a bottle neck for hardware implementation. This problem is very serious in compression of satellite images on-board, wh