Discrete cosine transform based processing framework for indexing, decomposition and compression of biospeckle data

Abstract

Biospeckle analysis is a useful tool for nondestructive characterization of different samples in diverse areas such as agriculture, engineering, biomedical imaging, medicine, and many more. In this paper, we present a single transform based framework to address three major aspects, viz. numerical quantification, sub-band decomposition and compression of biospeckle data. The discrete cosine transform (DCT) has been advantageously used to develop foundation for all three processing paradigms. First, an efficient method for numerical quantification of biospeckle activity using DCT has been developed. The salient features of the proposed strategy include its extreme simplicity, low complexity, and high accuracy. Furthermore, DCT has also been used in conjunction with the traditional processing methods to decompose and analyze the biospeckle activity associated with different sub-bands. The analysis allows isolation of different physiological and biological processes responsible for overall biospeckle behavior inside the specimen. DCT based image compression strategy to efficiently process the biospeckle data is also reported. Compression enables a way to faithfully process the biospeckle images with lower number of reconstruction coefficients and allows efficient storage and transmission. Performance evaluation and comparison of the proposed technique with the existing methods has been performed in both theoretical and experimental domains. To demonstrate practical utility of the proposed processing strategies blood coagulation process has been studied.