Compression-based self-organizing recognizer PRDC-CSOR with preliminary application to EO-image analysis

This paper introduces a new data analyzer, the compression-based self-organizing recognizer PRDC-CSOR, with a preliminary application to an EO-image. PRDC-CSOR is an application of the authors´ previously proposed pattern representation scheme using data compression (PRDC). Contrary to the traditional statistical model based recognition system methods, PRDC-CSOR constructs itself by using incoming data only. The basic tool, compressibility, is an approximation of the Kolmogorov complexity K(x) defined on an individual text x as a countermeasure to the Shannon entropy H(X) defined on an ensemble X. Due to this feature, a highly automatic self-organizing recognition system becomes possible as shown in this paper.