Garmash V. Information technology of images filtering and saving in challenging lighting conditions

Object of research - the processes of filtration and storage of digital images taken in low-light conditions; the purpose is to enhance the quality of digital images produced in low-light conditions through the development of information technology based methods of multiresolution analysis, bilateral filtering, methods and models, taking into account the characteristics of the image quality, optimal interpolation filter; used methods of multiresolution analysis to improve the filtration method of halftone images, linear algebra, and digital image processing for filtering of color images, the wavelet transform and numerical methods for reducing blocking-effect on halftone images, statistical analysis process and the results of tests, object-oriented programming in Microsoft Visual Studio Express 2010 for the implementation of information technology in the form of a software product; theoretical results: for the first time developed a method of image filtering to reduce blocking-effect that uses a dyadic wavelet transform and optimal interpolation filter which improves the visual quality of the image; improved method of filtering noise in halftone images based on bilateral filter due to its integration with the wavelet - filter bank, thus improving the signal/noise ratio by an average of 0.1 - 1 dB; improved method "Non - Local Means" to filtering noise in color images through the use of models of color component separating space YCrCb, thus improving the signal/noise ratio by an average of 0.5 - 2.5 dB and, in general, the visual quality of the images. The practical results - by improving the known methods and the application of new methods proposed in the thesis, built and implemented as software a new information technology filtering digital images, which has improved the quality of images taken in low-light conditions. Developed and implemented methods for noise filtering and reduction of blocking-effect of digital images, obtained under conditions of low light level in the form of algorithms. The proposed information technology can be used for video surveillance solutions for the problem of formation of long-term storage and archive solutions for the problems of automated video analysis. The degree of implementation - the main results are implemented in enterprises and institutions, "Vinnitsia Chamber of Commerce and Industry - " private enterprise "Anklyz" private enterprise "TVA - Service" as well as in the educational process of automatics and information - measuring equipment Vinnitsa National Technical University. Sphere of use - digital imaging systems, video surveillance systems.