NNIC-neural network image compressor for satellite positioning system

Pavel Danchenko, Feodor Lifshits, Itzhak Orion, Sion Koren, Alan D. Solomon, Shlomo Mark

Research output: Contribution to journalArticlepeer-review

4 Scopus citations


We have developed an algorithm, based on novel techniques of data compression and neural networks for the optimal positioning of a satellite. The algorithm is described in detail, and examples of its application are given. The heart of this algorithm is the program NNIC-neural network image compressor. This program was developed for compression color and grayscale images with artificial neural networks (ANNs). NNIC applies three different methods for compression. Two of them are based on neural networks architectures-multilayer perceptron and kohonen network. The third is based on a widely used method of discrete cosine transform, the basis for the JPEG standard. The program also serves as a tool for determining numerical and visual quality parameters of compression and comparison between different methods. A number of advantages and disadvantages of the compression using ANNs were discovered in the course of the present research, some of them presented in this report. The thrust of the report is the discussion of ANNs implementation problems for modern platforms, such as a satellite positioning system that include intensive image flowing and processing.

Original languageEnglish
Pages (from-to)622-630
Number of pages9
JournalActa Astronautica
Issue number8-9
StatePublished - Apr 2007
Externally publishedYes


  • Artificial neural networks (ANN)
  • Data compression
  • NNIC
  • Satellite positioning


Dive into the research topics of 'NNIC-neural network image compressor for satellite positioning system'. Together they form a unique fingerprint.

Cite this