Essential Data Elements: Extraction and Recovery

In tradition, equivalent data matrices or ordered data sets must be in the same size. We propose a new computing method which breaches the tradition and builds equivalence between data sets in different sizes. By the new method, essential data elements can be extracted from a data set or data source without loss of information. And, the full data set may be exactly recovered from its essential elements. Both extraction and recovery may be accomplished in real time. As such, the equivalence between data sets and their essential subsets can be validated by the invertible operations. This equivalence extends traditional concepts and methodologies in mathematics, data science and data processing engineering. It may greatly improve both veracity and velocity of data analytics. The new method may be extensively applied to image and video technologies, medical imaging, remote sensing, wireless communication and so on. It greatly improves efficiency of data acquisition, compression and visualization, especially in scalable applications. It makes revolutionary improvement than existing methods. Demo software and testing data are all downloadable at website http://qualvisual.net.