Abstract: In this paper, a new multi-sensor image fusion method that combines the source images by optimally weighting in terms of predetermined regions is proposed. The Artificial bee colony algorithm is employed to determine weights. The regions are produced by K-Means algorithm. The experiments are conducted with images for different applications. The proposed method is objectively and subjectively compared with well-known image fusion methods from the literature. Experimental results show that the method outperforms the other methods.

• Publication Date: 11-Oct-2015
• DOI: 10.15224/978-1-63248-064-4-01
• Views: 0
• Downloads: 0

Abstract: The speech recognition system and creation of program Software based on it is one of the important components of natural language computer modelling. For example, the popular Google Voice system, by which the word or sentence pronounced by a person is automatically written in textual form, in Google’s Search system field for given language by using already realized speech recognition systems. The trial Software program has been developed by us, which is able to solve the problems, as for the Google search system, also for YouTube, Yahoo, Bing, Yandex and Wikipedia in Georgian language. Also, it is possible to translate automatically Georgian words or sentences pronounced individually and get texts in the appropriate language by using Google Translate. Program management is performed with Georgian voice commands.

• Publication Date: 11-Oct-2015
• DOI: 10.15224/978-1-63248-064-4-02
• Views: 0
• Downloads: 0